Carboxylate Salt Bridge-Mediated Enamine Catalysis: Expanded Michael Reaction Substrate Scope and Facile Access to Antidepressant (R)-Pristiq

Article abstract of DOI:10.1002/adsc.201700801

We report broad guidance on how to catalyze enantioselective aldehyde additions to nitroalkene or maleimide Michael electrophiles in the presence of unprotected acidic spectator groups, e.g., carboxylic acids, acetamides, phenols, catechols, and maleimide

Full text of DOI:10.1002/adsc.201700801

Title: Carboxylate Salt Bridge Mediated Enamine Catalysis: Expanded
Michael Reaction Substrate Scope and Facile Access to
Antidepressant (R)-Pristiq
Authors: Thomas Nugent, Hussein Ali El Damrany Hussein, Shahzad
Ahmed, Foad Tehrani Najafian, Ishtiaq Hussain, Tony
Georgiev, and Mahmoud Khalaf Aljoumhawy
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201700801
Link to VoR: http://dx.doi.org/10.1002/adsc.201700801

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
Carboxylate Salt Bridge Mediated Enamine Catalysis: Expanded Michael Reaction
Substrate Scope and Facile Access to Antidepressant (R)-Pristiq
Thomas C. Nugent,* Hussein Ali El Damrany Hussein, Shahzad Ahmed, Foad Tehrani Najafian, Ishtiaq
Hussain, Tony Georgiev, Mahmoud Khalaf Aljoumhawy
Department of Life Sciences and Chemistry, Jacobs University Bremen, Campus Ring 1, 28759 Bremen
(Germany)
*e-mail: t.nugent@jacobs-university.de
KEY WORDS: organocatalysis, carboxylate salt bridge, Michael reaction, nitroalkene, maleimide,
peracetic acid
Graphical Content
Abstract: We report broad guidance on how to catalyze enantioselective aldehyde additions to
nitroalkene or maleimide Michael electrophiles in the presence of unprotected acidic spectator groups,
e.g., carboxylic acids, acetamides, phenols, catechols, and maleimide NH groups. Remarkably, these L-
threonine and L-serine potassium salt catalyzed reactions proceed even when the nucleophilic and
electrophilic Michael partners simultaneously contain acidic spectator groups. These findings begin to
address the historical non-compatibility of enantioselective catalytic reactions in the presence of acidic
moieties and simultaneously encroach on the reaction capabilities normally associated with cellular
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
environments. A carboxylate salt bridge, from the catalyst enabled enamine to the Michael electrophile, is
thought to facilitate the expanded Michael substrate profile. A practical outcome of these endeavors is a
new synthetic route to (R)-Pristiq, (-)-O-desmethylvenlafaxine, an antidepressant, in the highest yield
known to date because no protecting groups are required.
Michael reactions embody many different nucleophile/electrophile pairings making them good proving
grounds for probing and applying new catalytic methods. An exhaustively examined example is the
enantioselective addition of aldehydes to ortho-, meta-, or para-substituted--nitrostyrenes (Scheme 1).[1]
A large array of electron rich and poor aromatic substituents are compatible and excellent yield and ee
are noted. However, when a weakly acidic functional group (pKa= 0 to 12) is present, high level
achievement is restricted to -nitrostyrene substrates containing an ortho-OH or ortho-NHAc
substituent.[2,3] A substrate based ortho-directing effect, in the transition state, has been offered as a
plausible explanation.[2g] In short, despite the comprehensive study of aldehydic Michael additions to -
nitrostyrenes, only three examples are known with a coexisting acidic spectator group when an ortho-OH
or ortho-NHAc substituent is lacking. Those examples are restricted to meta- or para-based phenolic
alcohol substrates and involve the addition of acetaldehyde,[4] propanal,[5] or isobutyraldehyde.[6]
Carboxylate salt based enamine catalysis is known but not widely employed,[7,8] and we speculated that
its application could overcome the non-compatibility of acidic spectator functional groups during
enantioselective catalysis, in particular for the Michael reaction. Here we show that threonine or serine
potassium salt catalysis: (i) far surpasses the lone catalysis examples employing 3- or 4-hydroxy--
nitrostyrenes;[4-6] (ii) is applicable to unreported and more acidic spectator groups, e.g., 3,4-catechols, 3-
or 4-positioned acetamide or carboxylic acid moieties; and (iii) allows both the Michael electrophile and
nucleophile to simultaneously contain an acidic spectator group. In total, these results bear out the
hypothesis that potassium salts of amino acids have broadened the substrate breadth for enantioselective
enamine catalysis and we propose that these expanded reaction capabilities are due to a carboxylate salt
bridge that assembles the starting materials.
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
Scheme 1. Enantioselective aldehyde additions to -nitrostyrenes containing acidic moieties.
Ignoring, temporarily, the challenge of coexisting acidic spectator groups, it is noteworthy that a smaller
number of reports show the addition of -branched aldehydes,[7-9] as opposed to linear aldehydes, to -
nitrostyrenes.[1] We have consequently focused on -branched aldehyde additions here, which lead to the
more difficult to form quaternary carbon based Michael products. Our initial investigations (Table 1)
focused on adding isobutyraldehyde, the benchmark -branched aldehyde substrate, and
cyclohexanecarboxaldehyde, documented as difficult to add.[7a-c,9a,b,e,f]
Figure 1. The potassium salts of threonine, serine, leucine, alanine, and aspartic acid derivatives were
screened.
We have previously shown that OtBu-L-Thr (Figure 1) is capable of adding isobutyraldehyde (4) to -
nitrostyrenes,[7c,e] and modified conditions there from have now permitted us to readily add
isobutyraldehyde to -nitrostyrenes with meta- or para-positioned carboxylic acid, acetamide, catechol
units, or phenolic OH moieties in good yield (70-86%) and excellent ee (94-97%), see structures 2a, b, d,
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
f, g, I, j of Table 1. For cyclohexanecarboxaldehyde additions, OtBu-L-Ser proved to, always, be the
optimal amino acid from those shown in Figure 1. For example when a carboxylic acid, acetamide,
catechol, or phenolic OH spectator group was present (Table 1, see products 2c, 2e, 2j, 2l), both the
yields (62-86%) and ees (90-95%) were dramatically higher under OtBu-L-Ser catalysis. For perspective,
cyclohexanecarboxaldehyde (5) has never been added in the presence of an acidic spectator group, and
excluding our earlier study[7c] and this one, its addition to simple -nitrostyrene (no acid groups present)
always required a ≥20 mol%[7a,b,9 a,b,e,f] catalyst loading and, in the best outcome, resulted in 51% yield
(80% ee) for the Michael product.[9e]
These initial results were rounded out by adding cyclopentanecarboxaldehyde, Table 1 product 2h, albeit
optimally with a silyl protected threonine catalyst: OTBDPS-L-Thr (Figure 1).[10] Except for compounds 2f
and 2i, all Table 1 products are new and have been fully characterized (Supporting Information). For the
formation of product 2i (para-OH) and 2f (ortho-OH), our starting material stoichiometry and catalyst
loading far exceed the previous findings.[2e,6,11]
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
Table 1. Quaternary carbon Michael product formation in the presence of acidic spectator functionality.[a,b]
[a]
For representative solvents, see the Experimental Section of the manuscript [b] The potassium salt of
[c]
[d]
the shown amino acid was used.
The ee was determined for the corresponding lactone.
Cyclopentanecarboxaldehyde (6) used.
The Table 1 Michael products (2) establish the broad applicability of enantioselective aldehyde addition to
-nitrostyrenes in the presence of mildly acidic functional groups. Of further importance, the first examples
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
of aldehyde addition to a -nitrostyrene substituted by a carboxylic acid, see products 2a-c, were realized;
and no racemic examples of the same reactions exist. Those results led us to pursue a higher level
challenge: the first Michael reaction in which both the electrophile and the nucleophile contain an acidic
moiety. To test this possibility and simultaneously examine stereogenic quaternary carbon formation, we
added a phenol containing nonsymmetrical -branched aldehyde (7) to 3-OH--nitrostyrene and
separately to maleimide[12,13] (Scheme 2). The produced Michael products (8 and 9) contain vicinal
quaternary-tertiary stereogenic centers and were obtained in high ee and good yield under remarkably
practical starting material stoichiometries and reasonable catalyst loadings (Scheme 2). Motivated by our
inability to separate diastereomers 9a/9b and the further application potential of these products, we
converted them into the separable dihydropyrroles 10a/10b and fully characterized them.
Scheme 2. Stereogenic quaternary carbon formation when both the nucleophile and electrophile contain
an acidic moiety. Formation of adjacent stereogenic centers.
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
In total (Schemes 1 and 2), these results are significant due to the lack of examples of enantioselective
catalytic reactions that tolerate the co-existence of acidic functional groups. In short, the demonstrated
acidic spectator groups are reminiscent of those allowed within cellular reaction environments, but not
chemical ones.
(±)-Venlafaxine (Scheme 3) is a widely prescribed anti-depressant whose HCl salt is marketed as Effexor.
The cytochrome P-450 metabolite thereof, (±)-O-desmethylvenlafaxine or Pristiq, has largely replaced the
sale of Venlafaxine because of its improved half-life and inhibitor potency (norepinephrine and dopamine
uptake).[14] (R)-Pristiq (Scheme 3) is known to be a more active antidepressant than racemic Pristiq and is
patent protected for that indication.[15] In aggregate, this family of compounds has a rich synthetic history
which is now briefly summarized before elaborating on how the application of the above outlined
methodology has allowed the highest yielding synthesis of (R)-Pristiq to date.
Enantioselective syntheses are not known for (R)-Pristiq but are known for Venlafaxine,[16] the best
among them in 25% overall yield.[16a] However none are industrially used. All syntheses of (R)-Pristiq rely
on a common strategy in which a starting material with a protected phenolic OH group is advanced to (±)-
Venlafaxine, resolved to (R)-Venlafaxine, and finally O-demethylated[17] to reveal the phenolic OH of (R)-
Pristiq. A resolution allows a maximum 50% yield and the best resolution of (±)-Venlafaxine employs di-p-
toluoyl-D-tartaric acid, providing (R)-Venlafaxine in 24% yield.[14c,15] The best available O-demethylation
procedures are high yielding (>80%) but require high energy reagents or high temperature, as exemplified
o
by treatment with: nBuLi/diphenylphosphine,[15] or thiolate, e.g. anhydrous sodium sulfide, at ≥145 C,[18]
making their industrial application feasible but of lower economic value. Finally, in 2009 a three step
synthesis of (±)-Pristiq from 4-methoxyphenylacetonitrile was reported in 26% overall yield,[19] but to date
no one has described a method allowing the resolution of racemic Pristiq to (R)- or (S)-Pristiq. In short, no
synthesis of (R)-Pristiq is able to circumvent the yield inefficient phenolic OH protect/deprotection and
resolution steps. What follows is our protection group free enantioselective synthesis of (R)-Pristiq which
holds potential as a scalable industrial synthesis.
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
Scheme 3. The first enantioselective synthesis of (R)-Pristiq.
Inexpensive 4-hydroxybenzaldehyde (11) was converted to 4-OH--nitrostyrene (1f) using catalytic FeCl3
in 76% yield (Scheme 3).[20] The next reaction step, enantioselective Michael addition of
cyclohexanecarboxaldehyde, required non-trivial catalyst, solvent, and molarity screening beyond our
initial phenolic substrate optimizations. For example, the potassium salt of OtBu-L-Thr provided excellent
yield and ee when adding isobutyraldehyde to ortho-, meta-, and para-OH--nitrostyrenes, but attempts to
broaden the aldehyde substrate scope to a vastly more hindered -branched aldehyde, specifically
cyclohexanecarboxaldehyde, resulted in non-practical outcomes. Emblematic of those challenges was
the addition of cyclohexanecarboxaldehyde to para-OH--nitrostyrene, which resulted in the Prisitq
Michael product (2j) in 71% yield and 79% ee under catalysis with the potassium salt of OtBu-L-Thr.
We reasoned that this mediocre result originated from the increased steric congestion encountered in the
transition state when adding sterically hindered cyclohexanecarboxaldehyde vs isobutyraldehyde, and by
extension we speculated that reducing the steric bulk of the catalyst might overcome this problem. After
considering the likely enamine and transition state factors, Scheme 4, we noted that the OtBu-L-Ser
enamine would have a reduced energetic penalty for rotation about its C2-C3 bond vs the OtBu-L-Thr
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
enamine (Scheme 4, right panel), and this, in turn, would reduce the steric interaction of the -OtBu group,
of the catalyst, with the cyclohexane ring of cyclohexanecarboxaldehyde, while still enforcing high
enamine facial selectivity. These presumptions appear to be borne out in the final product profile for
product 2j, 86% yield with 95% ee, under OtBu-L-Ser catalysis (10 mol%). These optimized conditions
additionally proceed with a practical 1.0 to 1.5 stoichiometry for the starting materials: 4-OH--
nitrostyrene/cyclohexanecarboxaldehyde.
The next challenge was selective conversion of aldehyde 2j to the corresponding formate ester 12.
Excellent comprehensive reviews of Baeyer-Villiger oxidation are available,[21] but trends for the migratory
aptitude of non-aromatic aldehydes are not explicitly discussed. Fortunately, literature examples
published after those seminal reviews do show that secondary and tertiary carbon migration can be
favored over hydrogen migration for aliphatic aldehydes when using mCPBA.[22] With that precedent, we
converted aldehyde 2j into formate ester 12 in the presence of mCPBA, but the corresponding hydrolysis
product, alcohol 13, was always noted in ~10% yield. Accompanying the mCPBA with a phosphate based
salt did not change the reaction outcome. Of greater detriment was our inability to removal the m-
chlorobenzoic acid by product from alcohol 13 via chromatography or the application of aqueous acid-
base work-up procedures. These considerations and the fact that mCPBA is not used on an industrial
scale prompted our investigation of industrially acceptable 36-40% peracetic acid in acetic acid,[23] which
is commercially available. In the event, we again noted formate ester 12 and alcohol 13 formation. After
considerable investigation we found it of operational convenience and beneficial (no loss in yield, see
methods A versus B in the Supporting Information, Section 9) to use the crude Baeyer-Villiger product for
the subsequent hydrolysis step. Employing this two-step method, aldehyde 2j was consistently converted
to alcohol 13 in no less than 47% overall yield. The two-step overall yield did not improve when using
mCPBA.
Transformation of alcohol 13 directly into (R)-Pristiq was trivial. After nitro group reduction with Pd/C and
H2, and without work-up or further modification, excess aqueous formaldehyde was added and the
reaction re-pressurized under hydrogen. This one-pot procedure allowed an 80% yield of (R)-Pristiq after
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
chromatographic purification. Despite the obvious nature of this sequence, we were surprised to find no
other examples in which a nitro group has been converted directly to a dimethylamine when using
hydrogen as the main reductant. In total, this synthesis represents: (i) the first enantioselective synthesis
and (ii) at 25% overall yield, from 4-hydroxybenaldehyde, the highest overall yield to date for (R)-Pristiq
(Scheme 3).
Regarding critical points within the catalytic cycle of the developed Michael reaction, the enamine
transition state conformation and the assembled salt-bridge, carboxylate to potassium cation to nitro
group (see transition state in Scheme 4), have been previously elaborated on via earlier DFT studies
within an earlier manuscript of ours, albeit for a maleimide electrophile.[7e] Among other cations, e.g.,
lithium, sodium, rubidium, and cesium, the potassium cation is critical for high yield and selectivity. The
importance of the potassium cation was further underscored when the reaction of 1f to 2j (Scheme 3)
resulted a in 15% yield of 2j, after 30 h, in the presence of equal molar quantities of 18-C-6 (10 mol% -
same as the catalyst loading). It should be further noted that catalysis with only OtBu-L-Ser, i.e., without
KOH, resulted in no reaction after 48 h.
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
Scheme 4. Probable OtBu-L-Ser potassium salt catalytic cycle for cyclohexanecarboxaldehyde addition to
4-OH--nitrostyrene (1f) to give Pristiq Michael product 2j, also see Scheme 3. Right panel: OtBu-L-Ser
versus OtBu-L-Thr enamine steric considerations.
Based on the available information, one conclusion is that the shown substrates were not previously
viable because the acidic spectator groups negatively impacted vital proton exchange equilibriums
required within the catalytic cycle and/or these substrates were not sufficiently electrophilic enough to
react. It is important to note that enamine based Michael reactions have been exhaustively examined and
are overwhelmingly reported with steric based aminocatalysts and less frequently with amino acids (albeit
with no base added). By contrast, performing these reactions with the carboxylate salt of an amino acid
catalyst allows the catalyst enabled enamine and the Michael electrophile to assemble. This forces the
reacting carbon centers within bonding distance proximity with much greater frequency. This point alone
is likely the most influential one, and what separates these findings from the earlier ones. If correct, it is
remarkable that the catalyst’s carboxylate moiety can achieve productive equilibriums of the suggested
transition state assemblies despite the presence of stoichiometric quantities of acidic spectator groups
which will participate in non-productive protonation/deprotonation equilibriums.
Conclusion:
Carboxylate salt mediated enamine catalysis has resulted in a dramatic broadening of the acceptable
Michael reaction nucleophiles and electrophiles to those containing an acidic spectator group. The
demonstrated reaction capabilities expand chemical reactions into those formerly reserved for cellular
environments. This advance is potentially explainable through a carboxylate salt hinge that facilitates
highly congested and ordered transition states to reliably assemble despite the presence of competing
acidic spectator groups, and the findings should be extendable to other reaction types. Regarding the
product breadth, the method is tolerant of sterically hindered aldehyde additions and stereogenic
quaternary carbon based Michael products can be formed. Finally, this opening report will likely spur the
investigation of a broader number of amino acid catalysts, e.g., cysteine analogs, and perhaps more
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
importantly optimization of the amino acid protecting groups from O-t-Bu to, among others, pivaloyl or
benzoyl based esters, etc.
Experimental Section:
179 pages of experimental descriptions and characterization data are available in the Supporting
Information.
Generic Michael reaction procedure: To a screw cap vial was added KOH (10-60 mol%), catalyst (5.0-
20.0 mol%), and then a solvent/cosolvent (see next paragraph). This was stirred for 2-3 min before
adding the aldehyde (1.3-3.0 equiv). The reaction vessel was stirred for no more than 5 min and then a -
nitrostyrene derivative or maleimide (1.00 equiv) was added. Please refer to individual compound
description for the specific starting material stoichiometries, solvents, and reaction times. All reactions
were performed at 26 oC.
Solvents: It is important to note that each -nitrostyrene category (phenol versus acetamide versus
carboxylic acid substituents) required a solvent optimization, which more often than not was related to the
solubility of the -nitrostyrene. Examination of single solvents revealed either incomplete reactions (low
conversion) or low ee. These deficits were overcome by screening for synergistic combinations of those
same solvents. Solvent/cosolvent screening from MeOH, EtOH, acetone, THF, EtOAc, CH3CN, CH2Cl2,
tBuOMe, toluene, and n-pentane allowed the optimal binary solvent combination to be determined.
Representative examples follow for products 2a, 2d, 2j, 8. See the Supporting Information for even
greater detail.
(S)-3-(3,3-dimethyl-1-nitro-4-oxobutan-2-yl)benzoic acid (2a)
Reaction time= 30 h, KOH (MW= 56.11, 60.0 mol%, 0.31 mmol, 17.39 mg), O-tBu-L-threonine (MW=
175.23, 10.0 mol%, 0.05 mmol, 8.8 mg), tBuOMe/MeOH (8.5:1.5 vol ratio, 2.0 mL, 0.26 M),
isobutyraldehyde (MW= 72.11, 2.0 equiv, 1.04 mmol, 75.00 mg, 94.9 µL), (E)-3-(2-nitrovinyl)benzoic acid
(MW= 193.16, 1.00 equiv, 0.52 mmol, 100.4 mg). Rf = 0.41 (EtOAc/petroleum, 3:7) containing 3-4 drops
of acetic acid. Silica gel chromatography provided the pure product as a white sticky solid (MW= 265.26,
101 mg, 0.38 mmol, 73% yield).
95% ee: Chiralcel OD-H chiral HPLC column, iPrOH/n-heptane (20:80), the n-heptane was a 0.30 vol%
AcOH solution of n-heptane (vol/vol), flow rate = 0.8 mL/min, λ = 254 nm, tmajor= 22.5 min, tminor= 20.6 min.
1H NMR (400 MHz, DMSO-d6) (ppm): 0.88 (s, 3H), 1.04 (s, 3H), 3.97 (dd, J = 11.7, 3.6 Hz, 1H), 5.00 (dd,
J = 13.7, 3.9 Hz, 1H), 5.24 (t, J = 13.2 Hz, 1H), 7.46 (t, J = 7.6 Hz, 1H), 7.61 (d, J = 7.6 Hz, 1H), 13.08
(bs, 1H), 7.85 (m, 2H), 9.58 (s, 1H). 13C NMR (100 MHz, DMSO-d6) (ppm): 18.7, 20.1, 47.3, 48.0, 75.9,
128.5, 128.6, 129.9, 130.9, 133.4, 137.1, 167.2, 205.0. IR (ATR mode):Vmax = 2964, 2918, 2854, 1689,
1554, 1376, 1294. MS (EI, negative ion mode), m/z (relative intensity): 158 (100%). HRMS (ESI-TOF)
m/z: [M – H+]- Calculated for C13H14NO5:264.0877; Found: 264.0880.
(S)-N-(4-(3,3-dimethyl-1-nitro-4-oxobutan-2-yl)phenyl)acetamide (2d)
Reaction time= 20 h, T= 26 ºC, KOH (MW= 56.11, 13.0 mol%, 0.05 mmol, 2.80 mg), O-tBu-L-threonine
(MW= 175.23,10.0 mol%, 0.04 mmol, 7.00 mg), EtOAc/acetone 3.5:1.5 (2.0 mL, 0.20 M),
isobutyraldehyde (MW= 72.11, 3.00 equiv, 0.80 mmol, 57.68 mg, 73.00 µL), (E)-N-(4-(2-
nitrovinyl)phenyl)acetamide (MW= 206.20, 1.00 equiv, 0.40 mmol, 82.0 mg). Note, this is the only reaction
for which the equiv of isobutyraldehyde were not optimized. Rf = 0.31 (MeOH/CH2Cl2, 02:98). Silica gel
chromatography provided the pure product as a pale yellow sticky solid (MW = 278.30, 91 mg, 0.33
mmol, 83% yield).
97% ee: Chiralcel OD-H chiral HPLC column, iPrOH/n-heptane (15:85), flow rate = 1.0 mL/min, λ = 254
nm, tmajor= 46.8 min, tminor= 25.3 min.
1H NMR (400 MHz, CDCl3) (ppm): 0.96 (s, 3H), 1.09 (s, 3H), 2.10 (s, 3H), 3.73 (dd, J = 4.1 Hz, 11.5 Hz,
1H), 4.66 (dd, J = 4.2, 13.0 Hz, 1H), 4.82 (dd, J = 11.6, 12.9 Hz, 1H), 7.10 (d, J = 8.5 Hz, 2H), 7.45 (d, J =
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
8.5 Hz, 2H), 8.07 (s, 1H), 9.48 (s, 1H). 13C NMR (100 MHz, CDCl3) (ppm): 18.9, 21.6, 24.4, 48.0, 48.4,
76.4, 120.0, 129.6, 130.9, 138.1, 169.0, 204.5. IR (ATR mode): Vmax= 1721, 2720, 1668, 3191, 3119,
1516 cm-1. MS (EI, positive ion mode), m/z (relative intensity): 160 (100%). HRMS (ESI-TOF) m/z: [M
+ H+]+ Calculated for C14H19N2O4: 279.1339; Found: 279.1337.
(S)-1-(1-(4-hydroxyphenyl)-2-nitroethyl)cyclohexanecarbaldehyde (2j)
o
Reaction time= 22 h, T= 26 C, KOH (MW= 56.11, 0.15 equiv, 0.545 mmol, 30.6 mg), O-tBu-L-serine
(MW= 161.20, 10 mol%, 0.363 mmol, 58.6 mg), EtOAc/n-pentane (3:1 volume ratio, 4.5 mL, 0.8 M),
cyclohexanecarboxaldehyde (MW= 112.17, 1.50 equiv, 5.45 mmol, 611 mg), 4-hydroxy-β-nitrostyrene
(MW= 165.15, 1.00 equiv, 3.63 mmol, 600 mg). Rf= 0.27 (EtOAc/petroleum ether, 15:85). Silica gel
chromatography provided the pure product as a light yellow viscous oil which sometimes solidifies (MW=
277.32 MW, 3.14 mmol, 871 mg, 86% yield).
95% ee: Chiralcel OD-H chiral HPLC column, i-PrOH/Heptane (15:85), flow rate = 0.8 mL/min, λ = 254
nm, tminor= 15.6 min and tmajor= 29.8 min.
1H NMR (400 MHz, CDCl3) (ppm): δ 1.05-1.27 (m, 4H), 1.37 (dt, J= 12.6, 3.7, 1H), 1.55-1.71 (m, 3H),
1.88 (m, 1H), 2.06 (m, 1H), 3.48 (dd, J= 11.0, 4.8 Hz, 1H), 4.70 (dd, J= 13.0, 4.8 Hz, 1H), 4.76 (dd, J=
12.9, 11.1 Hz, 1H), 5.29 (bs, 1H), 6.74 (d, J= 8.6 Hz, 2H), 6.98 (d, J= 8.6 Hz, 2H), 9.53 (s, 1H). 13C NMR
(100 MHz, CDCl3) (ppm): δ 22.7, 22.8, 25.2, 29.9, 31.1, 49.9, 51.6, 76.4, 115.7, 126.8, 130.4, 155.6,
207.9. IR (ATR mode): Vmax= 3400, 2924, 2854, 1716. (ESI-TOF)-MS m/z: [M-H+]- Calcd for C15H19NO4
276.1235; Found 276.1241. MS (EI), m/z (relative intensity): 107.96, 133.93, 121.01, 163.84, 169.88.
(S)-2-((S)-2,5-dioxopyrrolidin-3-yl)-4-(4-hydroxyphenyl)-2-methylbutanal (8)
Reaction time= 60 h, T= 3 oC, KOH (MW= 56.11, 8.0 mol%, 0.04 mmol, 2.2 mg), O-tBu-L-threonine (MW=
175.23, 5.0 mol%, 0.025 mmol, 4.4 mg), CH2Cl2 (0.5 mL, 1.0 M). 4-(4-hydroxyphenyl)-2-methylbutanal (7)
(MW= 178.23, 1.3 equiv, 0.650 mmol, 115.8 mg), maleimide (MW= 97.1, 1.0 equiv, 0.50 mmol, 48.5 mg).
Rf(major)= 0.31, Rf(minor)= 0.35 (EtOAc/petroleum ether, 1:1). Careful silica gel chromatography
provided the major diastereomer as a white solid (MW= 275.30, 95.1 mg, 0.345 mmol, 69% yield).
96% ee: Chiralpak IA chiral HPLC column, EtOAc/n-heptane (50:50), flow rate = 1.0 mL/min, λ = 254 nm,
tmajor= 12.9 min and tminor= 17.5 min.
1H NMR (DMSO-d6, 400 MHz) δ (ppm): 0.98 (s, 3H), 1.56-1.80 (m, 2H), 2.21 (dt, J= 4.56, 12.77 Hz, 1H),
2.37 (dt, J= 5.24, 12.80 Hz, 1H), 2.62 (dd, J = 5.80, 18.21 Hz, 1H), 2.75 (dd, J = 9.29 Hz, 18.17 Hz, 1H),
3.48 (dd, J = 5.89, 9.06 Hz, 1H), 6.65 (d, J = 8.26, 2H), 6.96 (d, J = 8.27 Hz, 2H), 9.60 (s, 1H). 13C NMR
(DMSO-d6, 100 MHz) δ (ppm): 13.5, 28.5, 31.2, 36.5, 45.6, 50.1, 115.1, 129.1, 131.5, 155.5, 177.8,
179.9, 204.4. IR (ATR, cm-1): ν= 3142, 1710, 1516. MS (EI), m/z (relative intensity, negative mode):
273.78. HRMS (ESI-TOF) m/z: [M-H+]- Calculated for C15H16NO4: 274.1085; Found: 274.1073
Acknowledgements: We thank Panyapon Sudkaow and Vishal S. Dodke for recording some spectra
and initial reaction investigations. This work was generously supported by the Deutsche
Forschungsgemeinschaft (German Science Foundation) under award no. NU 235/7-1, and Jacobs
University Bremen. We are very thankful for low and high resolution MS measurements provided by
Professor Dr. Nikolai Kuhnert.
References:
[1] a) S. Sulzer-Mossé, A. Alexakis, Chem. Commun. 2007, 3123–3135; b) N. Mase, Enamine catalysis of
Michael reactions, in Science of Synthesis, Asymmetric Organocatalysis (Eds.: B. List, K. Maruoka),
Georg Thieme Verlag, Stuttgart, Germany, 2012, vol. 1, 135-216.
[2] For ortho-OH examples, see: a) B.-C. Hong, P. Kotame, C.-W. Tsai, J.-H. Liao, Org. Lett. 2010, 12,
776-779; b) D. Enders, C. Wang, X. Yang, G. Raabe, Adv. Synth. Catal. 2010, 352, 2869–2874; c) D. B.
Ramachary, R. Sakthidevi, Org. Biomol. Chem., 2010, 8, 4259–4265; d) D. Lu, Y. Li, Y. Gong, J. Org.
Chem. 2010, 75, 6900–6907; e) B.-C. Hong, P. Kotame, J.-H. Liao, Org. Biomol. Chem., 2011, 9, 382–
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
386; f) D. B. Ramachary, M. S. Prasad, R. Madhavachary, Org. Biomol. Chem. 2011, 9, 2715–2721; g) D.
B. Ramachary, R. Sakthidevi, K. S. Shruthi Chem. Eur. J. 2012, 18, 8008–8012.
[3] For ortho-NHAc examples, see: a) S. Heo, S. Kim, S.-G. Kim, Tetrahedron Lett. 2013, 54, 4978-4981;
b) Y. Lee, S. G. Kim, J. Org. Chem. 2014, 79, 8234-8243; c) M. Yu, S.-G. Kim, Tetrahedron Lett. 2015,
56, 4159-4162; A.-B. Xia, X.L. Zhang, T. Wang, X.-H. Du, D.-Q. Xu, Z-Y. Xu, New J. Chem. 2015, 39,
5088-5091.
[4] For the enzymatic addition of acetaldehyde to 3-OH or 4-OH--nitrostyrene, respectively in a 74%
yield (49% ee) and 59% yield (51% ee), see: a) E. Zandvoort, E. M. Geertsema, B.-J. Baas, W. J. Quax,
G. J. Poelarends, Angew. Chem., Int. Ed. 2012, 51, 1240–1243; b) E. M. Geertsema, Y. Miao, P. G.
Tepper, P. de Haan, E. Zandvoort, G. J. Poelarends Chem. Eur. J. 2013, 19, 14407–14410.
[5] For a proline analog (10 mol%) catalyzed addition of propanal (3.0 equiv) to 4-OH--nitrostyrene in
high yield and ee, see Scheme 5 of: I. Sagamanova, C. Rodríguez-Escrich, I. G. Molnꢀr, S. Sayalero, R.
Gilmour, M. A. Pericꢁs ACS Catal. 2015, 5, 6241−6248.
[6] For the aminocyclohexane catalyzed (20 mol%) addition of isobutyraldehyde (5.0 equiv) to 4-OH- or 4-
OH,3-MeO--nitrostyrene in high yield and ee, see Table 3 of: A. Ying, S. Liu, Z. Li, G. Chen, J. Yang, H.
Yan, S. Xu Adv. Synth. Catal. 2016, 358, 2116 – 2125.
[7] a) M. Yoshida, A. Sato, S. Hara Chem. Commun. 2008, 6242–6244; b) M. Yoshida, A. Sato and S.
Hara, Org. Biomol. Chem., 2010, 8, 3031–3036; c) Nugent, T. C.; Shoaib, M.; Shoaib, A. Org. Biomol.
Chem. 2011, 9, 52-56; d) M. Yoshida, E. Masaki, H. Ikehara, S. Hara, Org. Biomol. Chem. 2012, 10,
5289–5297; e) Nugent, T. C.; Sadiq, A.; Bibi, A.; Heine, T.; Zeonjuk L. L.; Vankova, N.; Bassil, B.S.
Chem. Eur. J. 2012, 18, 4088-4098.
[8] See the Supp Info within: H. Uehara, C. F. Barbas, III Angew. Chem. Int. Ed. 2009, 48, 9848–9852.
[9] a) N. Mase, R. Thayumanavan, F. Tanaka, C. F. Barbas III Org. Lett., 2004, 6, 2527-2530; b) J. Wang,
H. Li, B. Lou, L. Zu, H. Guo, W. Wang Chem. Eur. J. 2006, 12, 4321–4332; c) M. P. Lalonde, Y. Chen
and E. N. Jacobsen, Angew. Chem., Int. Ed., 2006, 45, 6366–6370; d) S. H. McCooey and S. J. Connon,
Org. Lett., 2007, 9, 599–602; e) J.-F. Bai, X.-Y. Xu, Q.-C. Huang, L. Peng, L.-X. Wang Tetrahedron Lett.
2010, 51, 2803–2805; f) Y. Qiao, L. Zhang, S. Luo, J.-P. Cheng Synlett 2011, 495–498; g) L. Tuchman-
Shukron, S. J. Miller, M. Portnoy Chem. Eur. J. 2012, 18, 2290–2296.
[10] Among the tested catalysts, OTBDPS-L-Thr was not commercially available. For its synthesizes, see:
C. Palomo, M. Oiarbide, A. Landa J. Org. Chem. 2000, 65, 41-46.
[11] As stated earlier, ortho-substrates are thought to participate directly in the Michael transition state,
see ref. 2g.
[12] Maleimide Michael reactions have been previously examined, see: a) J. Flores-Ferrándiz, B. Fiser, E.
Gómez-Bengoa, R. Chinchilla, Eur. J. Org. Chem. 2015, 1218–1225; b) P. Vízcaíno-Milla, J. M. Sansano,
C. Nájera, B. Fiser, E. Gómez-Bengoa Synthesis 2015, 47, 2199–2206; c) A. Avila, R. Chinchilla, E.
Gómez-Bengoa, C. Nájera Eur. J. Org. Chem. 2013, 5085–5092.
[13] A recent advance for maleimide Michael reactions has been reported, see: C. E. Grînenfelder, J. K.
Kisunzu, H. Wennemers, Angew. Chem. Int. Ed. 2016, 55, 8571–8574;
[14] a) E. A. Muth, J. A. Moyer, J. T. Haskins, T. H. Andree, G.E. M. Husbands Drug Dev. Res. 1991, 23,
191-199; b) K. J. Klamerus, K. Maloney, R. L. Rudolph, S. F. Sisenwine, W. J. Jusko, S. T. Chiang, J.
Clin. Pharmacol. 1992, 32, 716-724; c) T. P. Jerussi, C. H. Senanayake, N. N. Bhongle (Sepracor Inc.),
Patent No. US 20040180952 A1, Sep. 16, 2004.
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700801
Advanced Synthesis & Catalysis
[15] T. P. Jerussi, C. H. Senanayake, N. N. Bhongle (Sunovion Pharmaceuticals Inc.), Patent No. US
9072700 B2, July 7, 2015.
[16] H. M. L. Davies, A. Ni Chem. Commun. 2006, 3110–3112, (b) R. Bhuniya, S. Nanda Tetrahedron
Lett. 2012, 53, 1990–1992, (c) S. P. Chavan, K. P. Pawar, S. Garai RSC Adv. 2014, 4, 14468–14470.
[17] Alternative protecting groups have been examined but offer no improvement, see for example: L. A.
Turski, A. Stoit, C. G. Kruse, S. Vader, M. Th. M. Tulp (Solvay Pharmaceuticals), US 20090005455 A1,
01 January 2009.
[18] K. Pospisilik, L. Thijs (Synthon IP Inc.), US 7491848 B2, 17 February 2009.
[19] B. Lal, V. Gund, V. K. Pandian (Calyx Chemicals Pharmaceuticals Ltd.), WO 2009084037 A2, 09 July
2009.
[20] S. Jalal, S. Sarkar, K. Bera, S. Maiti, U. Jana, Eur. J. Org. Chem. 2013, 22, 4823–4828.
[21] For example, see: (a) M. Renz, B. Meunier Eur. J. Org. Chem. 1999, 737-2750; (b) G.-J. ten Brink, I.
W. C. E. Arends, and R. A. Sheldon Chem. Rev. 2004, 104, 4105-4123.
[22] For secondary and tertiary carbon migration vs. hydrogen migration during Baeyer-Villiger oxidations,
see: (a) C. Margot, D. P. Simmons, D. Reichlin, D. Skuy Helv. Chim. Acta 2004, 87, 2662-2684; (b) H.
Leuser, S. Perrone, F. Liron, F. F. Kneisel, P. Knochel Angew. Chem., Int. Ed. 2005, 44, 4627–4631; (c)
R. Chowdhury, S. K. Ghosh Tetrahedron: Asymmetry 2011, 22, 1895–1900.
[23] For example, see: L. T. Boulton, D. Brick, M. E. Fox, M. Jackson, I. C. Lennon, R. McCague, N.
Parkin, D. Rhodes, G. Ruecroft Org. Process Res. & Dev. 2002, 6, 138-145.
This article is protected by copyright. All rights reserved.