Gelatin protein-mediated direct Aldol reaction

Article abstract of DOI:10.1002/hlca.201300380

Gelatin protein was found to catalyze the aldol reaction between cyclohexanone and different aromatic aldehydes under mild reaction conditions. The aldol additions carried out in DMSO at 37 yielded the addition products with moderate diastereoselectivitie

Full text of DOI:10.1002/hlca.201300380

5
74
Helvetica Chimica Acta – Vol. 97 (2014)
Gelatin Protein-Mediated Direct Aldol Reaction
a
1
a
1
a
b
by Dennis Kꢀhbeck ), Jꢀrgen Bachl ) ), Eva-Maria Schçn ) ), Vicente Gotor-Fern a´ ndez ), and David
a
c
Dꢁaz Dꢁaz* ) )
a) Universitꢀt Regensburg, Institut fꢁr Organische Chemie, Universitꢀtsstrasse 31, DE-93053
Regensburg
(
)
phone: þ 49-941-943-4373; fax: þ 49-941-943-4121; e-mail: David.Diaz@chemie.uni-regensburg.de)
b
Universidad de Oviedo, Departamento de Quꢂmica Org a´ nica e Inorg a´ nica, Avenida Juli a´ n Claverꢂa
s/n, ES-33071 Oviedo
c) IQAC-CSIC, Jordi Girona 18-26, ES-08034 Barcelona
Gelatin protein was found to catalyze the aldol reaction between cyclohexanone and different
aromatic aldehydes under mild reaction conditions. The aldol additions carried out in DMSO at 378
yielded the addition products with moderate diastereoselectivities favoring the syn isomers. Appropriate
control experiments demonstrated the activity of the protein in the aldol reaction. The kinetic study of
the model reaction between 4-nitrobenzaldehyde and cyclohexanone established a first-order rate
ꢂ3 ꢂ1
constant k ¼ (7.4 ꢀ 0.5) ꢁ 10 h . Moreover, the scale-up of the process was successfully achieved at 1-g
scale in a yield comparable to that in small scale.
Introduction. – Since the first report in 1872 [1], the aldol reaction between two
carbonyl compounds has become one of the most powerful CꢂC bond-forming
methods in organic synthesis. During the last decades, a number of different systems,
including small molecules, polymers, enzymes, and antibodies, have been reported to
serve as efficient catalysts for this important reaction [2 – 5].
In this context, and motivated by the production of value-added products from
sustainable resources, we decided to explore the potential of gelatin as a natural
organocatalyst for CꢂC bond formation. Gelatin is a biodegradable protein (50 –
100 kD) obtained either by acid (type A) or alkaline (type B) processing of collagen,
which is the main component of connective tissues and the most abundant protein in
mammals. The average amino-acid composition often follows the pattern Gly-Pro-X
and Gly-X-Hyp, where X is any other amino acid. In general, commercial samples
contain Gly (ca. 23%), Pro (ca. 12%), and Hyp (ca. 11% of the total amino acid
content). It should be noted that these values might vary depending on the source of
the material and processing method [6]. The fact that some amino acids and peptides
have been employed in different catalytic studies [2][7] prompted us to investigate the
potential ability of gelatin to promote some organic reactions. This protein has been
traditionally used as a general ingredient in food, cosmetic, pharmaceutical, and
photographic industries [8], and only recently as a reducing ligand in the preparation of
metal nanoparticles [9].
1)
Both authors have contributed equally to this work.
ꢃ 2014 Verlag Helvetica Chimica Acta AG, Zꢁrich

Helvetica Chimica Acta – Vol. 97 (2014)
575
As a result of our studies, we have recently reported the addition of nitro alkanes to
aldehydes (Henry or nitroaldol reaction) catalyzed by gelatin and collagen proteins in
both aqueous and organic media [10]. We have expanded here the potential of the
gelatin protein, reporting the results obtained from the direct aldol reaction between
cyclohexanone and different aromatic aldehydes in the presence of commercial gelatin
under mild reaction conditions.
Results and Discussion. – Based on our previous experience and recent reports
[
(
10][11], the reaction between 4-nitrobenzaldehyde (1a, 0.1 mmol) and cyclohexanone
2) at physiological temperature was chosen as the model aldol reaction to study the
activity of the gelatin protein (i.e., gelatin from porcine skin type-A (PSTA)).
Preliminary screening of the reaction conditions (see Exper. Part) revealed that 10 mg
of gelatin and tenfold molar excess of 2 with respect to 1a were optimal to obtain the
corresponding addition product 3a in reasonable yields. Higher loadings did not
significantly improve the results. For instance, the product yield obtained in the model
reaction using 10 equiv. of 2 was ca. 62%, whereas the use of 20 equiv. afforded the
product in ca. 64% yield. In contrast, the use of only 5 equiv. of 2 decreased the yield to
38%. Thus, further studies on the substrate scope were carried out using the mentioned
pre-optimized conditions (Table 1).
Regarding the solvent, DMSO was found suitable for carrying out a comparative
study. For instance, the product yield obtained in DMSO was ca. fourfold higher than in
Table 1. Gelatin Protein-Catalyzed Aldol Reactiona)
b
c
Entry
R
1
Solvent
DMSO
Yield ) of 3 [%]
dr ) [syn/anti]
1
2
3
4
5
6
7
8
9
4-NO ꢂC H
1a
1a
1a
1b
1c
1d
1e
1f
1g
1h
1i
62
14
13
54
15
52
24
20
34
10
5
3.9/1.0
1.3/1.0
1.5/1.0
4.1/1.0
1.5/1.0
4.2/1.0
3.6/1.0
2.3/1.0
2.8/1.0
2.2/1.0
2.5/1.0
2
6
4
H O
H O )
2
2
d
3-NO ꢂC H
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
2
6
4
4
2-NO ꢂC H
2
6
4-NCꢂC H
6
4
Pyridin-2-yl
Ph
4-BrꢂC H
6
4
1
1
0
1
4-MeꢂC H
6
4
4-MeOꢂC H
6
4
a) Reaction conditions: 1 (0.1 mmol), 2 (1.0 mmol), gelatin PSTA (10 mg), solvent (0.5 ml), 378, and 7 d
b
1
at 250 rpm. ) Yields ( H-NMR) that correspond to the average values of two independent experiments
c
1
(
standard deviation, ꢀ 2%). ) Diastereomer ratio syn/anti determined by H-NMR analyses of two
independent experiments. Relative configurations were assigned by comparison with the data reported in
the literature. ) Reaction carried out in the presence of Bu NBr (13 mg, 0.04 mmol).
d
4

5
76
Helvetica Chimica Acta – Vol. 97 (2014)
Scheme. Control Experiment in the Absence of Gelatin Protein
H O under the same conditions (Entries 1 and 2). In contrast to our previous
2
observations with the Henry (nitroaldol) reaction [10], the use of a phase transfer co-
catalyst such as Bu NBr [12] did not improve the yield of the gelatin-mediated aldol
4
reaction in H O (Entries 2 and 3).
2
It is important to highlight that appropriate control experiments confirmed the
catalytic activity of the protein. For instance, only a tiny amount of the aldol product 3a
was detected when the model reaction between 1a and 2 was carried out either in
DMSO or in H O (pH ca. 4 – 5) for 7 d in the absence of gelatin (Scheme). In
2
agreement with our previous studies on the Henry reaction [10], we could also exclude
any catalytic effect on the aldol reaction possibly caused by other minor components
(ca. 10%) such as metal and ash impurities in the gelatin samples [9a][13]. The aldol
product was obtained in only 4% yield when the model reaction of 1a and 2 was
performed only in the presence of metal ions as Lewis acids at the concentration
ꢂ6
ꢂ1
2þ
reported to be present in gelatin samples (i.e., [metal] ꢁ 10 mol l : 3.92 (Co ); 5.0
2
þ
3þ
2þ
2þ
(
Cu ); 9.42 (Fe ); 2.62 (Pd ); 9.92 (Ni )) [13]. Most importantly, although the
specific mechanism of gelatin catalysis remains unclear, a simple acid or base catalysis
was ruled out by a) controlling the pH of the reaction medium (pH ca. 5 – 5.5), b) the
poor yields of the blank experiments (vide supra), and c) almost identical results
obtained using gelatin samples with different isoelectric points (i.e., PI gelatin type-
A ꢃ 6; PI gelatin type-B ꢄ 5), in agreement with our previous observations [10].
We further studied the gelatin-mediated reaction between different aldehydes 1a –
1i and 2 (Table 1). Aromatic aldehydes with strongly or moderately deactivating groups
(
i.e., electron-withdrawing substituents) were smoothly converted to the corresponding
b-hydroxy ketones in modest yields (Entries 1, 4, 5, and 6). The results also revealed the
influence of the substitution position on the product yield, indicating a significant
decrease with the ortho-substituted aldehyde 1c (Entries 1, 4, and 5). On the other
hand, low yields were obtained with pyridine-2-carbaldehyde 1e (Entry 7), benzalde-
hyde (1f, Entry 8), and in the cases of aromatic aldehydes bearing weakly or
moderately activating groups (i.e., electron-donating substituents, i.e., 1g – 1i; En-
tries 9 – 11). As observed with other aldol-like reactions, these yields could be improved
by increasing the reaction time and/or the temperature [10]. Unfortunately, aliphatic
aldehydes such as isovaleraldehyde were not converted after 7 d under the reported
conditions.
Interestingly, a modest syn diastereoselectivity was found in all cases (i.e., ca. two-
to fourfold of syn vs. anti), which is an uncommon feature in this reaction [2][3]. The
diastereoisomeric ratio (dr) was reduced either in H O, with orto-substituted aromatic
2
aldehydes or with activated aromatic aldehydes. As expected due to the mild reaction

Helvetica Chimica Acta – Vol. 97 (2014)
577
conditions, no potential by-products such as dehydrated or self-condensation products
were observed.
In accordance with our previous studies [10], the direct use of collagen also
afforded the desired aldol product (e.g., 1a (0.1 mmol), 2 (1.0 mmol), collagen (10 mg),
DMSO (0.5 ml), 7 d, 378: yield, 46%, dr (syn/anti), 3.4 :1.0).
It is also worth mentioning that the reaction can be scaled-up with yields com-
parable to those of the small scale (i.e., 1a (1.13 g, 7.5 mmol), 2 (7.75 ml, 75 mmol),
PSTA gelatin (0.75 g), DMSO (38 ml), 7 d, 378; ca. 58% yield). In terms of recycling,
gelatin was quantitatively recovered and reused in at least two consecutive runs with
maintenance of stereoselectivity. Nevertheless, a major catalyst deactivation was ob-
served in the third run (Fig. 1). This is presumably due to inefficient molecular desorp-
tion from the catalyst, instead of decomposition, as indicated by a visible gradual color
change (from light-yellow to intense yellow-orange) and similar FT-IR spectra of the
washed-dried catalyst after each run. Either acetone or EtOH could be used to precipi-
tate the protein from DMSO, which was further separated by centrifugation. Alter-
natively, direct extraction of the aqueous solutions with AcOEt allows simple isolation
and subsequent reuse of the protein-containing aqueous solution in the next cycle.
Despite the chiral secondary structure of gelatin and collagen proteins, chiral high-
performance liquid chromatography of the reaction mixtures revealed almost
negligible enantioselectivity (i.e. < 5% ee for the major diastereoisomer, syn) during
the aldol reaction. This lack of enantioselectivity has been also reported with other
proteins or biopolymers when used as catalysts in aldol-like reactions [10][11][14]. In
contrast to the results reported with other biocatalysts [15], reducing the temperature
to 208 did not cause any increase of the enantiomeric selectivity.
The first-order kinetics analysis of the model reaction between 1a and 2 established
ꢂ3 ꢂ1
a slow rate constant k ¼ (7.4 ꢀ 0.5) ꢁ 10
h
(Fig. 2). It is worthwhile to mention that
the morphology and/or physical state of a protein catalyst may also play a significant
Fig. 1. a) Recycling experiment for the model reaction between 1a (7.5 mmol) and 2 (75 mmol) catalyzed
by PSTA gelatin (0.75 g) in DMSO (38 ml) at 378 for 7 d. b) Gradual coloring and apparent clogging of
gelatin. I: Gelatin before 1st run; II: washed-dried gelatin after 3rd run.

5
78
Helvetica Chimica Acta – Vol. 97 (2014)
Fig. 2. First-order kinetics plot of the model aldol reaction between 1a and 2 catalyzed by gelatin. C ,
1
Final concentration, at infinite time; C , concentration at given time t; C , initial concentration, at t ¼ 0.
t
0
2
R ¼ 0.98. Inset: Evolution of the reaction over time
role in the kinetics of aldol-like reactions under heterogeneous or semi-heterogeneous
conditions [10].
Conclusions. – In summary, we have demonstrated that CꢂC bond-formation
through direct aldol reaction between cyclohexanone and different aromatic aldehydes
can be promoted by commercial gelatin in organic medium under mild conditions. The
corresponding addition products were obtained in modest yields governed by first-
order kinetics. In general, modest syn-diastereoselectivity and complete chemo-
selectivity were also observed, demonstrating the applicability of the catalytic system in
a g-scale experiment. Moreover, the protein could be recovered at the end of the
reaction and reused in at least two consecutive cycles without major loss of the catalytic
activity. Finally, we would like to emphasize that these and our previous investigations
indicate that, although edible gelatin is not a competitive catalyst for this type of
reactions from a synthetic perspective, the finding that it can mediate to some extent
(even under aqueous conditions) the transformation of carbonyl compounds occuring
in numerous foods or cosmetic formulations could be significant from a metabolic point
of view.
Financial support of this work by Universitꢀt Regensburg (D. D. D., Fçrderlinie C des Finanziellen
Anreizsystems fꢁr Drittmitteleinwerbung) is gratefully acknowledged. D. D. D. thanks the Deutsche
Forschungsgemeinschaft (DFG) for the Heisenberg Professorship Award.
Experimental Part
Materials. Anal.-grade solvents and commercially available reagents were purchased from
SigmaꢂAldrich or TCI Europe, and used as received. Milli-Q H O was used for the experiments in aq.
2
solns. Gelatin porcine skin type-A (PSTA) was purchased from SigmaꢂAldrich (CAS 9000-70-8; Cat.

Helvetica Chimica Acta – Vol. 97 (2014)
579
No. G2500-100G; Batch No. 128K0066; Type A, derived from acid-cured tissue; ca. 300 Bloom; 79%
protein content by Buiret) and used without further purification. Collagen was purchased from
SigmaꢂAldrich (CAS 9007-34-5; Cat. No. C9879-1G; Batch No. 061M7015V; Type A, derived from
bovine achilles tendon). A detailed description of the most important features of both gelatin and
collagen can be found in the Supporting Information of our previous work focused on the nitroaldol
(
Henry) reaction catalyzed by these proteins [10].
Methods. TLC: Fluorescent-indicating plates (Al sheets precoated with silica gel 60 F 254; Merck);
visualization by the use of the phosphomolybdic acid as stain soln. and UV light (254 nm). HPLC: Varian
9
20-LC chromatograph (column, Phenomenex Lux Cellulose-2, 4.6 ꢁ 250 mm, 5 mm; eluent, heptane/
i
PrOH 70 :30; flow rate, 1.0 ml/min; l 254 nm). FT-IR Spectra: Diamond ATR (attenuated total
reflection) accessory (Golden Gate). H-NMR Spectra: Bruker Avance 300 spectrometer, at 258.
1
1
Yields were determined by H-NMR analyses of the crude product in CDCl with Ph CH (1 ml of a
3
2
2
0
.1m stock soln.) as internal standard. The results were confirmed by a second experiment using directly
N,N-dimethylacetamide (9.2 ml, added with a Hamilton syringe) as internal standard. Thus, possible
concentration variations of the stock soln. of Ph CH in CDCl could be detected and the values cross-
2
2
3
1
checked. Relative configurations were assigned by comparison with H-NMR data reported in [16].
1
For kinetics calculations, the H-NMR analyses of the mixtures were performed in the presence of
Ph CH (0.1 mmol) as internal standard. Each exper. point was the average of at least two independent
2
2
experiments. Among various kinetics models, the straight line shown in the kinetics plot show best-fit of
the first-order model (i.e., [cyclohexanone] ꢃ [aldehyde]).
NMR and IR spectra can be obtained directly upon request from the authors.
Preliminary Optimization Experiments. a) Screening of Cyclohexanone (2) equivalents. The model
reaction between 1a (15.1 mg, 0.1 mmol) and 2 (5, 10, and 20 equiv.) was carried out in DMSO (0.5 ml) at
3
6
78 for 7 d in the presence of PSTA gelatin (10 mg). That the aldol product 3a was obtained in 38, 62, and
4% yield, resp.
b) Screening of Catalyst Loading: Table 2 compiles the results of selected experiments for the
reaction between 1a (0.1 mmol) and 2 (2 mmol) in DMSO (0.5 ml) under different conditions of solvent
temp., reaction time, and catalyst loading.
General Procedure for Gelatin-Catalyzed Aldol Reaction. Cyclohexanone (2; 104 ml, 1.0 mmol) was
added in one portion to a 4-ml screw cap vial containing 4-nitrobenzaldehyde (1a; 15 mg, 0.1 mmol),
gelatin PSTA (10 mg), and DMSO (0.5 ml). The mixture was stirred (250 rpm) for 7 d at 378. The
reaction was quenched by the addition of AcOEt (1 ml) and EtOH (1 ml), and subsequently the
precipitated catalyst was filtered. The filtrate was rinsed three times with AcOEt (3 ꢁ 1 ml), and the
combined org. phases were washed with H O (2 ꢁ 5 ml) and brine (5 ml), dried (Na SO ), filtered, and
2
2
4
evaporated under reduced pressure to afford the crude product. Yield and diastereoisomer ratio (syn/
1
anti) were determined by H-NMR analyses of the crude product in CDCl (300 MHz) using Ph CH
3
2
2
(
1 ml of a 0.1m stock soln.) as internal standard after complete workup. Relative configurations were
assigned by comparison with the spectroscopic data reported in the literature. For instance, in the model
Table 2. Initial Screening of Catalyst Loadinga)
Entry
Solvent
Temp. [8]
Time [d]
Catalyst loading [mg]
Yield of 3a [%]
1
2
3
4
5
6
7
8
DMSO
DMSO
DMSO
25
25
25
25
37
37
25
37
14
18
14
21
7
7
21
7
0
5
10
0
0
0
2
27
62
2
7
5
H O
2
H
2
O
H O/Bu NBr
2
4
H O
5
10
2
14
2
H O
2
a) The use of more than 10 mg of PSTA renders the workup more laborious.

5
80
Helvetica Chimica Acta – Vol. 97 (2014)
reaction between 1a and 2, the anti diastereoisomer was clearly identified by a doublet at 4.85 ppm (J ¼
8
.3, 1 H), whereas the syn diastereomer displayed the doublet at 5.41 ppm (J ¼ 2.4, 1 H). Typically, for
recycling experiments, acetone or EtOH (1 ml per 2 mg of catalyst) was added in order to precipitate the
gelatin from DMSO. The protein was subsequently separated by centrifugation (10 min,
3
800 rpm)ꢂwashing with AcOEt (2 ml)-centrifugation cycles. The obtained residue was finally dried
under vacuum before the next catalytic cycle.
REFERENCES
1] A. Wurtz, Bull. Soc. Chim. Fr. 1872, 17, 436.
2] B. M. Trost, C. S. Brindle, Chem. Soc. Rev. 2010, 39, 1600, and refs. cit. therein.
3] T. D. Machakewski, C.-H. Wong, Angew. Chem., Int. Ed. 2000, 39, 1352, and refs. cit. therein.
4] R. Mahrwald, ꢄModern Methods in Stereoselective Aldol Additionꢅ, 1st edn. Wiley-VCH,
Weinheim, 2013.
[
[
[
[
[
[
5] K. Fesko, M. Gruber-Khadjawi, ChemCatChem 2013, 5, 1248.
6] R. N. R. M. Hafidz, C. M. Yaakob, I. Amin, A. Noorfaizan, Int. Food Res. J. 2011, 18, 813; R.
Schrieber, H. Gareis, ꢄGelatine Handbook: Theory and Industrial Practiceꢅ, 1st edn., Wiley-VCH,
Weinheim, 2007.
[
7] K. Rohr, R. Mahrwald, Org. Lett. 2012, 14, 2180; D. Deng, P. Liu, B. Ji, W. Fu, L. Li, Catal. Lett. 2010,
1
37, 163; Z. Jiang, H. Yang, X. Han, J. Luo, M. W. Wong, Y. Lu, Org. Biomol. Chem. 2010, 8, 1368; M.
Amedjkouh, Tetrahedron: Asymmetry 2007, 18, 390; D.-S. Deng, J. Cai, Helv. Chim. Acta 2007, 90,
14; T. Nagamine, K. Inomata, Y. Endo, L. A. Paquette, J. Org. Chem. 2007, 72, 123; N. Mase, Y.
1
Nakai, N. Ohara, H. Yoda, K. Takabe, F. Tanaka, C. F. Barbas III, J. Am. Chem. Soc. 2006, 128, 734;
A. C o´ rdova, W. Zou, P. Dziedzic, I. Ibrahem, E. Reyes, Y. Xu, Chem. – Eur. J. 2006, 12, 5383.
8] A. Imeson, ꢄFood stabilisers, thickeners and gelling agentsꢅ, Wiley-Blackwell, Chichester, 2010.
9] a) H. Firouzabadi, N. Iranpoor, A. Ghaderi, Org. Biomol. Chem. 2011, 9, 865; b) H. Firouzabadi, N.
Iranpoor, A. Ghaderi, J. Mol. Catal. A: Chem. 2011, 347, 38; c) S.-Z. Kang, T. Wu, X. Li, J. Mu,
Colloids Surf., A 2010, 369, 268.
[
[
[
10] D. Kꢁhbeck, B. B. Dhar, E.-M. Schçn, C. Cativiela, V. Gotor-Fern a´ ndez, D. Dꢂaz Dꢂaz, Beilstein J.
Org. Chem. 2013, 9, 1111.
[11] D. Kꢁhbeck, G. Saidulu, K. R. Reddy, D. Dꢂaz Dꢂaz, Green Chem. 2012, 14, 378.
[12] L. Mhamdi, H. Bohli, Y. Moussaoui, R. ben Salem, Int. J. Org. Chem. 2011, 1, 119.
[13] C. V.-L. Bray, F. Jiang, X.-F. Wu, J.-B. Sortais, C. Darcel, Tetrahedron Lett. 2010, 51, 4555.
[14] S. E. Milner, T. S. Moody, A. R. Maguire, Eur. J. Org. Chem. 2012, 3059; M. S. Humble, P. Berglund,
Eur. J. Org. Chem. 2011, 3391; E. Busto, V. Gotor-Fern a´ ndez, V. Gotor, Chem. Soc. Rev. 2010, 39,
4504; Q. Wu, B.-K. Liu, X.-F. Lin, Curr. Org. Chem. 2010, 14, 1966; J. Fan, G. Sun, C. Wan, Z. Wang,
Y. Li, Chem. Commun. 2008, 3792.
[
[
15] H.-H. Li, Y.-H. He, Z. Guan, Catal. Commun. 2011, 12, 580.
16] S. Paladhi, J. Das, P. K. Mishra, J. Dash, Adv. Synth. Catal. 2013, 355, 274; Z.-B. Xie, N. Wang, G.-F.
Jiang, X.-Q. Yu, Tetrahedron Lett. 2013, 54, 945; R. Pedrosa, J. M. Andr e´ s, R. Manzano, P.
Rodrꢂguez, Eur. J. Org. Chem. 2010, 5310; A. Ricci, L. Bernardi, C. Gioia, S. Vierucci, M. Robitzer,
F. Quignard, Chem. Commun. 2010, 46, 6288.
Received October 22, 2013