Biomimetic synthesis of bis-α-substituent pyrrolidine alkaloids based on a proposed biosynthetic pathway

Article abstract of DOI:10.1016/j.tetlet.2016.03.104

A possible biosynthetic pathway of bis-α-substituent pyrrolidine alkaloids was proposed. Based on the proposed biosynthetic pathway, six pyrrolozoxazine alkaloids and one N-alkyl-5-hydroxymethyl-pyrrole-2-carbaldehyde alkaloid were synthesized. In a mixture of acetic acid and triethylamine, condensation of d-fructose and d-amino acids produced pyrrolozoxazine alkaloids. Replacing d-amino acids with tyramine can afford pyrrolezanthine - a N-alkyl-5-hydroxymethyl-pyrrole-2-carbaldehyde alkaloid.

Full text of DOI:10.1016/j.tetlet.2016.03.104

Tetrahedron Letters 57 (2016) 2219–2221
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Biomimetic synthesis of bis-a-substituent pyrrolidine alkaloids
based on a proposed biosynthetic pathway
⇑
Zhi Cao, Yueqing Li, Shisheng Wang, Bo Tang, Xiuhan Guo, Liu Wang, Weijie Zhao
School of Pharmaceutical Science and Technology, State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian 116024, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
A possible biosynthetic pathway of bis-a-substituent pyrrolidine alkaloids was proposed. Based on the
proposed biosynthetic pathway, six pyrrolozoxazine alkaloids and one N-alkyl-5-hydroxymethyl-pyr-
role-2-carbaldehyde alkaloid were synthesized. In a mixture of acetic acid and triethylamine, condensa-
tion of D-fructose and D-amino acids produced pyrrolozoxazine alkaloids. Replacing D-amino acids with
tyramine can afford pyrrolezanthine—a N-alkyl-5-hydroxymethyl-pyrrole-2-carbaldehyde alkaloid.
Ó 2016 Published by Elsevier Ltd.
Received 26 December 2015
Revised 27 March 2016
Accepted 31 March 2016
Available online 1 April 2016
Keywords:
Bis-a-substituent pyrrolidine alkaloids
Pyrroloxazine alkaloids
Pyrrolozoxazine
Biosynthetic pathway
Pyrrolozoxazine alkaloids 1–6 (Fig. 1) from natural world were
reported successively. Alkaloid 1 was found from Capparis spinosa
in 2010.1 Alkaloids 1–6 were isolated from the fruits of Morus alba2
in 2014. Alkaloids 2–5 with R-configuration at C-8 were also found
in the fermentation broth of an endophytic actinomycetes.3
with yields of 11% and 4%.13 These Letters offered us enlighten-
ment that pyrrolozoxazine alkaloids can be obtained from natural
compounds directly. We believed the biosynthesis of bis-
a-sub-
stituent pyrrolidine alkaloids should be related with
and primary amines.
D-fructose
Pyrrolezanthine (7),
a
N-alkyl-5-hydroxymethyl-pyrrole-2-car-
Proposed biosynthetic pathway of alkaloids includes ornithine
pathway, lysine pathway, anthranilic acid pathway, phenylala-
nine/tyrosine pathway, and tryptophan pathway. Among them,
baldehyde alkaloid, was isolated from Formosan Zanthoxylum sim-
ulans,4 mushroom Leccinum extremiorientale,5 Alhagi sparsifolia6
and Actinobacterium Jiangella gansuensis.7 All alkaloids in Figure 1
contain the same substructure of formyl pyrrole. The formyl pyr-
role and oxazine are regarded as important pharmacophore units.
Alkaloid 6 significantly inhibited pancreatic lipase activity.2 Acor-
tatarins A bearing formyl pyrrole group and oxazine ring could
inhibit reactive oxygen species production in high-glucose-stimu-
lated mesangial cells.8
ornithine pathway is the main origin of
alkaloids. But the biosynthetic pathway of increasing occurrence
of bis-
-substituent-pyrrolidine alkaloids, such as acortatarins,14
pollenopyrroside15 and capparisine1 remains unknown. Herein
we proposed a possible biosynthetic pathway of bis- -substituent
a-substituent-pyrrolidine
a
a
pyrrolidine alkaloids including pyrrolozoxazine alkaloids 1–6 and
Pyrrolezanthine 7.
The existence of such kind of alkaloids in nature is interesting
for many researchers. We have done the total synthesis of alkaloid
Maillard reaction of primary amine with D-fructose or D-glucose
yielded 5-hydroxymethyl-pyrrole-2-carbaldehyde unit,16 but it
has not been proven occurring in plants yet. We suggested that
1 in six steps with a 17.0% overall yield,9 starting from
L-alanine.
We have been committed to the synthesis of bis-
pyrrolidine alkaloids, and synthesized two bis-
a
a
-substituent
-substituent
bis-a-substituent-pyrrolidine alkaloids may come from a type of
Maillard reaction of
D-fructose and primary amine in plants as
pyrrolidine alkaloids, pollenopyrroside A and capparisine B, from
D
Figure 2.
-fructose in 14 steps.10 The first synthesis of alkaloid 1 was
The fact that alkaloids 2, 3, and 4 were first found in the process
of roasting chicory root without any chemical reagent12 showed
pyrrolozoxazine alkaloids can be obtained from natural com-
pounds directly.
reported in 1971 via roasting DL-alanine and D-glucose at 200–
250 °C.11 Alkaloids 2, 3, and 4 were first found in the process of
roasting chicory root without any chemical reagent.12 Alkaloids 5
and 6 were achieved from monosaccharide ester as side products
Applying Maillard reaction, 5-hydroxymethyl-pyrrole-2-car-
baldehyde unit can be obtained from
as Figure 2. The amine part of alkaloids 1–6 should be amino acids.
The condensation of -glucose and primary amines has been
D-fructose and amino group
⇑
Corresponding author. Fax: +86 0411 84701411.
E-mail address: Pharmscizhao@163.com (W. Zhao).
D
http://dx.doi.org/10.1016/j.tetlet.2016.03.104
0040-4039/Ó 2016 Published by Elsevier Ltd.

2220
Z. Cao et al. / Tetrahedron Letters 57 (2016) 2219–2221
OH
OH
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
HO
O
O
N
1
N
2
N
5
N
3
N
4
N
6
N
Pyrrolezanthine(7)
Figure 1. Reported structures of alkaloids 1–7.
Table 1
OH
O
R
N
O
OH
-4H2O
Optimization in the condensation of D-fructose and D-alanine
OH
R
NH2
HO
+
Entry Mole
ratioa
Solvent
T
(°C)
t
(h)
Yieldb
(%)
OH OH
D-fructose
1
2
3
4
5
6
7
1:1
1:1
1:1
1:1
1:1
1:1
1:1
Acetic acid
Acetic acid
Acetic acid
Acetic acid
80
50
50
80
80
80
80
8
8
18
8
8
8
2
Figure 2. Proposed biosynthesis of alkaloids with 5-hydroxymethyl-pyrrole-2-
carbaldehyde unit.
1.2
1.3
2.1
0
2.2
3.8
EtOH, 1.5 equiv acetic acid
Acetic acid + pyridine (2:1)
Acetic acid + triethylamine
(2:1)
Acetic acid + triethylamine
(2:1)
Acetic acid + triethylamine
(2:1)
Acetic acid + triethylamine
(2:1)
reported in DMSO and oxalic acid.17 This kind of conversion
occurred in acid environment. As Paal–Knorr pyrrole synthesis
was conducted in acetic acid, we tried the dehydration in acetic
8
8
1:2
1:3
1:5
1:5
1:6
80
80
80
80
80
8
8
8
8
8
8.2
acid from -fructose and alanine at 80 °C as Scheme 1. Although
D
alkaloids 1 was obtained, the yield was only 2% calculated with ala-
nine (entry 1 in Table 1). The side product 5-hydroxymethylfur-
fural (5-HMF) and large amount of formed caramel became a
barrier of desired alkaloids.
The changes of reaction conditions with lower temperature
(entry 2 in Table 1), longer incubation time (entry 3 in Table 1)
9
8.7
10
11
12
9.6
Acetic acid + triethylamine
(4:3)
Acetic acid + triethylamine
(4:3)
18.4
18.6
or D-fructose replaced by D-glucose (entry 4 in Table 1) did not
a
increase the yield. We cut down the usage of acetic acid (entry 5
in Table 1) and found that alkaloid 1 could not be generated. In
the mixture of acetic acid and pyridine, the yield had no obvious
increments (entry 6 in Table 1). Triethylamine replacing pyridine
as component solvent13 (volume ratio = 2:1) could increase the
Mole ratio was the mole of
-alanine to -glucose.
Yield was calculated with
D-alanine to
D-fructose, except that entry 4 was mole
of
D
D
b
D-alanine.
Table 2
Yields of alkaloids 1–6 and corresponding enantiomeric excess (ee)
yield (up to 3.8%). The yield was raised with the augment of D-fruc-
tose (1:1–1:5, entries 7–10 in Table 1). When the ratio of acetic
acid and triethylamine was changed to 4:3, the yield reached
18.4% (entry 11 in Table 1). The mixture solvent could drastically
reduce the side products, including caramel and 5-HMF. So corre-
Entry
Substrate
Product
Yield (%)
Eea (%)
1
2
3
4
5
6
7
8
Alkaloid 1
Alkaloid 2
Alkaloid 3
Alkaloid 4
Alkaloid 5
Alkaloid 6
Compound 1a
Alkaloid 1
18.4
16.7
15.1
15.7
18.5
21.2
18.4
13.8
50
94
96
79
95
93
30
50
D
D
D
D
D
D
-Alanine
-Valine
-Isoleucine
-Leucine
sponding amino acid and D-fructose (5 equiv) were dissolved in
acetic acid and triethylamine (volume ratio = 4:3), then stirred at
80 °C for 5–10 h (monitored by TLC). Alkaloids 2–6 were obtained
in the yield of 15.1–21.2% (Table 2).
-Phenylalanine
-Tyrosine
The 1H and 13C NMR spectra data of synthetic products 1–6
L-Alanine
were identical with those reported data for natural products.2 In
D-Alanine methyl ester
20
view of the specific rotations, the synthetic products 2–5 ([a]
a
D
Yield was calculated with amino acid or amino acid methyl ester.
20
À32 to À88) were consistent with reported data ([
a]
À32 to
D
À88). The products of this route were enantioenriched while the
product by Bu et al. was raceme.9 The specific rotations of alkaloid
alkaloids 1–6 were inferred as R-configuration at C-8 based on lit-
1 found in Capparis spinosa was À33.3 (c 0.05, MeOH),1 which was
erature3 and specific rotations of our final products.
20
in accordance with our final product ([
a]
À36 (c 0.27, MeOH)) of
D
The mechanism of the dehydration of
acids is proposed in Figure 3. First, -fructose and amino acids
D-fructose and amino
D
-alanine. Otherwise, the specific rotations of 1a with S-configura-
D
tion at C8, the product of
L-alanine was +38 (c 0.08, MeOH). So the
formed Schiff base B in acid. Second, the Schiff base could be iso-
merized to enamine C.17 By eliminating one molecule of H2O,
enamine C was converted to intermediate D.18 Then, the nitrogen
attacked the C-5 atom to form the pyrrole ring (E). Effect of elec-
tron pushing of nitrogen atom promoted the removal of 4-hydroxyl
group and the aromatization to pyrrolic compound (F). Finally, our
target products were formed via intramolecular esterification in
acid.
We established the ee of products by high performance liquid
chromatography (HPLC) analysis employing a Chiralcel OD-H
(250 mm  4.6 mm) column as Table 2. The fact that ee of alkaloids
1 and 1a were 50% and 30% elucidated the conditions of acetic acid
O
R
OH
O
R
O
O
OH
80
䰳, 8h
O
+
HO
N
H2N
OH Et3N, CH3COOH
OH OH
D-fructose
D-amino acid
1: R= Me; 2: R= isopropyl; 3: R=sec-butyl
1-6
4: R=isobutyl; 5: R= benzyl; 6: R= 4-hydroxybenzyl
Scheme 1. Biomimetic synthesis of alkaloids 1–6.

Z. Cao et al. / Tetrahedron Letters 57 (2016) 2219–2221
2221
O
N
OH
O
OH
OH
O
2
H2N
R
O
1
5
3
OH
OH
R
OH
+
HO
6
4
OH HN
R
OH
-H2O
OH
OH OH
D-fructose
HO
HO
OH
Amino acid
OH OH
B
OH OH
C
O
OH
OH
O
R
N
O
O
O
-H2O
OH HN
R
O
R
-H2O
-H2O
O
R
HO
HO
HO
N
E
O
N
HO
O
OH
D
H
F
H
G (1-6)
Figure 3. Proposed mechanism of the dehydration of D-fructose and amino acid.
Supplementary data
O
NH2
N
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2016.03.
104.
CH3COOH
Et3N
+
D-Fructose
HO
HO
Pyrrolezanthine
HO
Tyramine
References and notes
Scheme 2. Biomimetic synthesis of pyrrolezanthine.
1. Yang, T.; Wang, C.-H.; Chou, G.-X.; Wu, T.; Cheng, X.-M.; Wang, Z.-T. Food Chem.
2010, 123, 705–710.
2. Kim, S. B.; Chang, B. Y.; Hwang, B. Y.; Kim, S. Y.; Lee, M. K. Bioorg. Med. Chem.
Lett. 2014, 24, 5656–5659.
3. Wang, P.; Kong, F.; Wei, J.; Wang, Y.; Wang, W.; Hong, K.; Zhu, W. Marine Drugs
2014, 12, 477–490.
4. Yang, Y.-P.; Cheng, M.-J.; Teng, C.-M.; Chang, Y.-L.; Tsai, I.-L.; Chen, I.-S.
Phytochemistry 2002, 61, 567–572.
5. Choi, J.-H.; Ozawa, N.; Yamakawa, Y.; Nagai, K.; Hirai, H.; Kawagishi, H.
Tetrahedron 2011, 67, 6649–6653.
6. Zou, G. A.; Mansur, S.; Hu, S. C.; Aisa, H. A.; Shakhidoyatov, K. M. Chem. Nat.
Compd. 2012, 48, 635–637.
7. Han, L.; Gao, C.; Jiang, Y.; Guan, P.; Liu, J.; Li, L.; Xu, L.; Huang, X. J. Nat. Prod.
2014, 77, 2605–2610.
8. Zhao, Z.-F.; Zhou, L.-L.; Chen, X.; Cheng, Y.-X.; Hou, F.-F.; Nie, J. Chin. Med. J.
2013, 126, 1230–1235.
9. Bu, X.; Li, Y.; Liu, J.; Zeng, D.; Zhao, W. Chem. Nat. Compd. 2012, 48, 194–197.
10. Cao, Z.; Li, Y. Q.; Wang, S. S.; Guo, X. H.; Wang, L.; Zhao, W. J. Synlett 2015, 921–
926.
and triethylamine led to partial racemization of the stereocenter.
But for alkaloids 2–6 (except for alkaloid 4), the ee were 93–96%.
There were little enantiomers formed. Based on the Letter before,19
the formation of labile Schiff bases (B) might be the reason of
racemization. And the steric and electronic properties of R group
influenced the formation of labile Schiff bases and then affected
racemization. The partial racemization provided further evidence
that our mechanism via Schiff bases was reasonable.
Maillard reaction of amine and
afford bis- -pyrrolidine alkaloids in a mild condition. As amino
acid and -fructose are natural substance, this way may also be
the biosynthetic pathway of bis- -pyrrolidine alkaloids. To vali-
date the pathway is general or not, pyrrolezanthine 7 was synthe-
sized from -fructose and tyramine (Scheme 2). In a mixture of
D-glucose or D-fructose can
a
D
a
11. Shigematsu, H.; Kurata, T.; Kato, H.; Fujimaki, M. Agric. Biol. Chem. 1971, 35,
2097–2105.
D
12. Sannai, A.; Fujimori, T.; Kato, K. Agric. Biol. Chem. 1982, 46, 429–433.
triethylamine and acetic acid (volume ratio is 4:3), the yield could
reach to 39.7%.
´
13. Jeric´, I.; Šimici, L.; Stipetic´, M.; Horvat, S. Glycoconjugate J. 2000, 17, 273–282.
14. Tong, X.-G.; Zhou, L.-L.; Wang, Y.-H.; Xia, C.; Wang, Y.; Liang, M.; Hou, F.-F.;
Cheng, Y.-X. Org. Lett. 2010, 12, 1844–1847.
15. Guo, J. L.; Feng, Z. M.; Yang, Y. J.; Zhang, Z. W.; Zhang, P. C. Chem. Pharm. Bull.
2010, 58, 983–985.
16. (a) Kato, H.; Sonobe, H.; Fujimaki, M. Agric. Biol. Chem. 1977, 41, 711–712; (b)
Antal, M. J., Jr; Mok, W. S. L.; Richards, G. N. Carbohydr. Res. 1990, 199, 91–109.
17. Adhikary, N. D.; Kwon, S.; Chung, W.-J.; Koo, S. J. Org. Chem. 2015, 80, 7693–
7701.
18. (a) Román-Leshkov, Y.; Chheda, J. N.; Dumesic, J. A. Science 2006, 312, 1933–
1937; (b) Moreau, C.; Durand, R.; Razigade, S.; Duhamet, J.; Faugeras, P.;
Rivalier, P.; Ros, P.; Avignon, G. Appl. Catal. A 1996, 145, 211–224.
19. Brückner, H.; Justus, J.; Kirschbaum, J. Amino Acids 2001, 21, 429–433.
In summary, bis-
achieved from natural
a
D
-substituent pyrrolidine alkaloids were
-fructose and -amino acids (or tyramine)
D
in a solution of acetic acid and triethylamine in the yield of
15.1–21.2%. The proposed biosynthetic pathway is suitable for
pyrrolezanthine in a 39.7% yield. Proteinogenic amino acids are
always L-stereochemistry, whereas natural D-amino acid should
be the major biosynthetic precursor of pyrrolozoxazine alkaloids
in view of our biomimetic route.