Solar and Visible Light Assisted Peptide Coupling

Article abstract of DOI:10.1002/anie.202011510

Amino acid and peptide couplings are widely used in fields related to pharma and materials. Still, current peptide synthesis continues to rely on the use of expensive, water sensitive, and waste-generating coupling reagents, which are often prepared in mu

Full text of DOI:10.1002/anie.202011510

Angewandte
Research Articles
Chemie
How to cite:
International Edition: doi.org/10.1002/anie.202011510
German Edition: doi.org/10.1002/ange.202011510
1
2
3
4
5
6
7
8
9
1
Photochemistry
Hot Paper
2
3
4
Solar and Visible Light Assisted Peptide Coupling
Abhaya K. Mishra, Galit Parvari, Sourav K. Santra, Andrii Bazylevich, Ortal Dorfman,
Jonatan Rahamim, Yoav Eichen,* and Alex M. Szpilman*
5
6
7
8
9
10 Abstract: Amino acid and peptide couplings are widely used in
11 fields related to pharma and materials. Still, current peptide
12 synthesis continues to rely on the use of expensive, water
13 sensitive, and waste-generating coupling reagents, which are
14 often prepared in multi-step sequences and used in excess.
15 Herein is described a peptide coupling reaction design that
16 relies mechanistically on sun-light activation of a 4-dimethyl-
17 amino-pyridine–alkyl halide charge-transfer complex to gen-
18 erate a novel coupling reagent in situ. The resulting coupling
19 method is rapid, does not require dry solvents or inert
20 atmosphere, and is compatible with all the most common
21 amino acids and protecting groups. Peptide couplings can be
22 run on gram-scale, without the use of special equipment. This
23 method has a significantly reduced environmental and finan-
24 cial footprint compared to standard peptide coupling reactions.
25 Experimental and computational studies support the proposed
26 mechanism.
In fact, with the exception of DCC, coupling reagents are 10
typically much more expensive than the substrate amino-acid 11
monomers employed in the reaction. A second issue is the 12
high molecular weight of typical coupling reagents such as 13
HBTU
(O-(benzotriazol-1-yl)-1,1,3,3-tetramethyluronium 14
hexafluorophosphate, 379.25 gmolÀ1), BOP ((benzotriazolyl- 15
N-oxy)-tris-(dimethylamino)-phosphonium hexafluorophos- 16
phate, 442.28 gmolÀ1) and PyBOP ((benzotriazol-1-yloxy)- 17
tripyrrolidino-phosphonium
hexafluorophosphate, 18
520.39 gmolÀ1), which means that a large mass of waste is 19
generated per equivalent of product in coupling reactions that 20
rely on these reagents. Developing a practical peptide 21
coupling method, which relies on fewer, readily available, 22
and potentially recyclable reagents would therefore consti- 23
tute an important advance in accordance with the principles 24
of atom economy and step economy.[8] Such a reaction would 25
still have to meet the challenges of avoiding base-catalyzed 26
epimerization of the starting and product peptides, in addition 27
to requiring short reaction times and taking place under mild 28
27
28 Introduction
29
30
conditions.
29
With the increased use of synthetic peptides in medicine,
In recent years visible light photo catalysis has become 30
31 biology and functionalized materials,[1] inarguably peptide
32 coupling[2] and amide bond formation[3] reactions have
33 become some of the most important synthetic transforma-
34 tions. Peptide coupling reactions are now being carried out on
35 large scale in industry with solution-phase techniques gen-
36 erally being preferred for kilogram to ton scale preparation of
37 shorter peptides, while solid-phase techniques are preferred
38 for smaller scale and medium peptide length.[4] The prepara-
39 tion of longer peptides sometimes relies on a combination of
40 solution and solid-phase strategies due to cost and purifica-
41 tion issues.[4,5] Since the development of dicyclohexyl carbo-
42 diimide (DCC),[6] a large array of modern peptide coupling
43 reagents have been developed in order to ensure high yields
44 for the widest possible spectrum of amino acid combinations
45 (Scheme 1a).[2,7] The more complex of these reagents are
46 prepared in multiple step sequences and are often used in
47 excess. This constitutes a financial and environmental burden.
48
one of the most prolific subjects in chemistry by virtue of 31
taking place under mild conditions and relying on commonly 32
available light sources.[9] The majority of recent research has 33
been carried out using light emitting diodes (LEDs) as an 34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
49
[*] Dr. A. K. Mishra, Dr. S. K. Santra, Dr. A. Bazylevich, O. Dorfman,
J. Rahamim, Prof. Dr. A. M. Szpilman
Department of Chemical Sciences, Ariel University
4070000 Ariel (Israel)
E-mail: szpilman@ariel.ac.il
50
51
52
53
Dr. A. K. Mishra, Dr. G. Parvari, Prof. Dr. Y. Eichen
Schulich Faculty of Chemistry, Technion – Israel Institute of
Technology, 3200008 Haifa (Israel)
E-mail: yoav@ch.technion.ac.il
54
55
56
57
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under https://doi.org/10.
1002/anie.202011510.
58
59
Scheme 1. This research in perspective.
59
&&&&
ꢀ 2021 Wiley-VCH GmbH Angew. Chem. Int. Ed. 2021, 60, 2 – 8
These are not the final page numbers!

Angewandte
Research Articles
Chemie
1
2
3
4
5
6
7
8
9
inexpensive and convenient light source. However, solar light Results and Discussion
1
2
3
4
5
6
7
8
9
is still the cheapest, most reliable, and readily available source
of light available in many parts of the world, notably in many
developing countries. Surprisingly, while much effort is
currently being invested into finding ways of harnessing solar
energy for electricity generation and water splitting,[10] using
visible light harnessed from the Sun to perform synthetic
chemistry is still in its pioneering stages.[11,12]
Charge-transfer complexes have long been known to be
activated by visible light to afford radicals and ammonium
radical cations.[17] We proposed that if Steglich base that is,
DMAP (1)[18] would form a charge-transfer complex such as 3
with a suitable alkyl halide it would, under the influence of
visible light, be converted into iminium salt 7, which would be
in equilibrium with 8 in situ (Scheme 2a). Both species could
To date there have been few attempts to develop visible
10 light-assisted amide coupling reactions. Known reactions rely
11 on aldehydes[13] or thioacids[14] as starting materials (Sche-
12 me 1b). Aldehydes and thioacids are inherently less attractive
13 as starting materials for peptide synthesis than native amino
14 acids, as they require additional steps for preparation, and
15 amino-aldehydes are innately unstable. Presumably due to
16 these considerations and with the exception of the thioacid to
17 amide coupling reported by the Tan group,[14b] these methods
18 were not applied to peptide coupling. A single example of
19 solar light-assisted amide formation of carboxylic acids and
20 trialkylamines has been reported by our group (Sche-
21 me 1c).[15] However, the specifics of the mechanism of this
22 method invariably results in the loss of one of the amine alkyl
23 substituents. It also requires long reaction times and neither
24 of these characteristics are compatible with or applicable to
25 peptide synthesis using regular amino acids. An alternative
26 visible light mediated approach to dipeptides is a-oxidation of
27 amines to provide amides.[16]
act as a peptide coupling reagent. The light-assisted oxidation 10
of 3 to form an iminium species such as 7 is in agreement with 11
the earlier findings of Lautenberger and others.[17] In this 12
reaction design DMAP (1) would take on the additional roles 13
of a base, and of an acylation catalyst, as outlined in 14
Scheme 2a–c.
15
Reaction of 7 or 8 with a carboxylic acid would lead to the 16
formation of a hemiaminal ester 11 and one equivalent of 17
DMAP hydrohalide salt 10. Based on the unique electronic 18
properties of hemiaminals, particularly the electron with- 19
drawing effect of the nitrogen atom, we envisioned that 11 20
would function as an activated carboxylate ester.[19] If this 21
hypothesis would prove to be correct, a molecule of DMAP 22
(1) would then catalyze the coupling of the carboxylate with 23
an amino acid or amine to form an amide or a peptide bond in 24
the product 13 with 14 as the leaving group.
25
Gratifyingly, DMAP (1) forms charge-transfer complexes 26
with various alkyl halides. As discussed below bromotri- 27
28
Herein we report a visible to near UV light-assisted
chloromethane, the best reagent from the synthetic point of 28
29 peptide and amide coupling reaction which relies on inex-
view, forms a charge-transfer complex 3 with DMAP (1) as 29
30 pensive and partly recyclable, low molecular weight, reagents,
shown in Scheme 2a. This is evidenced by comparison of the 30
31 4-DimethylAmino-Pyridine (DMAP (1), 122.17 gmolÀ1) and
UV-VIS spectrum of the mixture with that of the sum of the 31
32 BrCCl3 (198.13 gmolÀ1, Scheme 1d). These reagents are both
individual reagents (Figure S1a). This is likely an example of 32
33 mild and are not significantly hazardous relatively to existing
halogen bonding. This complex was also found by computa- 33
34 reagents. The reaction times are short, the yields excellent
tion, in both single molecule gas-phase studies (MP4SDTQ/6- 34
35 even on gram scale, and no epimerization takes place under
31G(2df,p)) and under solvent simulation (dichloroethane, 35
36 the reaction conditions. Standard amino acid protecting
SMD continuum solvation model, m062x/6-31G(2df,p)). Us- 36
37 groups, including Fmoc, CBz, and Boc, are compatible with
ing the latter level of theory for thermodynamic calculations, 37
38 the reaction conditions as detailed below. Excess reagent may
a stationary point at local energy minimum was found for the 38
39 be recycled. No drying of the solvents or inert gas is needed.
complex between bromotrichloromethane and DMAP. This 39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
complex, DMAP···BrCCl3 (3, Scheme 2a), was found to be 40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Scheme 2. General reaction design. a) The calculated structure (MP4SDTQ/ 6-31G(2df,p)) for the charge-transfer complex (3) of DMAP with
BrCCl3, shows a close N-Br contact, dN-Br =2.72 ꢀ, and co-linearity of the N-Br-C atoms, aN-Br-C =179.98. Also shown is the calculated structure of
the first triplet state, 4, (MP4SDTQ/ 6-31G(2df,p)). b) Reaction of carboxylic acid 9 with reactive species 7 or 8 to provides activated ester 11.
57
58
57
58
59
59 c) Peptide coupling with amine 12 to provide peptide 13.
Angew. Chem. Int. Ed. 2021, 60, 2 – 8
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
1
2
3
4
5
6
7
8
9
slightly endergonic, DG =+ 4.3 kcalmolÀ1 at 298 K.[20] The
dependency of the energy of this complex on the N-Br
distance is provided in the SI and these data show that the
complex is indeed a bound state with a calculated energy
barrier for breaking into its components DMAP (1) and
BrCCl3 (2) of about 5 kcalmolÀ1. Thus, at room temperature
this complex should exist in rapid dynamic equilibrium with
its components.
character of this transformation (See the Computational
Section in the SI). These results also support the mechanistic
claim of dynamic breaking of the complex upon irradiation.
Attempts to optimize the triplet excited state complex with
a solvation model at the m062x/6-31G(2df,p) level of theory
produced a very similar structure geometrically, but failed to
produce a local energy minimum.
1
2
3
4
5
6
7
8
9
For the optimization study we purposely focused on the
relatively challenging coupling reaction with hindered amino
In the optimized gas-phase complex 3 (Scheme 2), the
10 nitrogen atom of the pyridine group of DMAP binds to the
11 bromine atom of BrCCl3, with a dN-Br = 2.72 ꢀ (MP4SDTQ/6-
12 31G(2df,p)). This distance is significantly shorter than the sum
13 of the vdW radii of the atoms (dN-Br = 3.40 ꢀ), a clear
acid L-isoleucine, in its Fmoc protected form (16), and the 10
carboxylic acid with glycine ethyl ester as the amine 11
component (Table 1). For convenience, the commercial 12
hydrochloride salt of glycine ethyl ester (17) was used. A 13
borosilicate glass flask was used for all the experiments. This 14
glass is characterized by a light transmittance cut-off at 15
around 330 nm (See Figure S2). The reactants 16 and 17 were 16
dissolved and reacted under light irradiation with DMAP (1/ 17
18a) and a variety of alkyl-halides 19 (Table 1, See Figure S3 18
for a picture of the reaction set-up). As alluded to, BrCCl3 was 19
found to be the superior charge-transfer acceptor reagent, in 20
accordance with above mentioned calculations. With BrCCl3 21
(10 equivalents in DCE, Table 1, Entry 1), and 2 equivalents 22
of glycine ethyl ester, the reaction was complete within 23
45 minutes, whereas 12 hours and 4 equivalents of glycine 24
ethyl ester hydrochloride (17) were required for the reaction 25
to reach completion with CCl4 (co-solvent with DCM, 26
Table 1, Entry 2) or CBr4 (Table 1, Entry 3). The optimized 27
conditions (Table 1, Entry 1) provide the desired dipeptide 28
Fmoc-Ile-Gly-OEt (20) in 79% isolated yield, after column 29
chromatography, on a 200 mg scale. Surprisingly, the reaction 30
could be run under air with no need for a dry or inert 31
atmosphere. Internal temperature typically reached 30–328C 32
at the end of the reaction (See Supporting Information for 33
À
14 indication of bonding. The C Br bond length, dC-Br = 1.97 ꢀ,
À
15 is within the calculation error of the C Br bond length in free
16 BrCCl3. The N-Br-C angle in the complex is aN-Br-C = 179.98.
17
In order to further clarify the impact of photochemical
18 activation of the charge-transfer complex, the structure of the
19 DMAP···BrCCl3 complex was optimized under the constraint
20 of a triplet state (MP4SDTQ/ 6-31G(2df,p)). The resulting
21 structure 4 (Scheme 2a) differs considerably from that of its
À
22 ground-state counterpart 3, and clearly indicates C Br bond
23 cleavage, in parallel to incipient bond formation between the
24 Br atom and the nitrogen atom of the pyridine ring in DMAP.
À
25 The C Br distance in this complex is dC-Br = 3.93 ꢀ, which
À
26 indicates incipient bond cleavage. The N Br distance is dN-
À
27 Br = 2.52 ꢀ, which indicates a considerable degree of N Br
28 bond formation and therefore transfer of the Br to the DMAP
29 molecule. The N-Br-C angle also changes considerably upon
30 photoexcitation, transforming the structure away from the
31 ground state linear complex, with a N-Br-C angle of aN-Br-C
32 179.98, into a bent structure, with a N-Br-C angle of aN-Br-C
=
=
33 74.98 for the excited state. Hirshfeld charge distribution
34 analysis (MP4SDTQ/6-31G(2df,p))[21] supports the radical
details about light flux).
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
35
36
37
Table 1: Optimization study and control experiments.
38
39
40
41
42
43
Entry
Light source
solvent
n
m,18
p,19
Time
Yield[b]
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
1
2
3
4
5
6
7
8
Sun
Sun
Sun
Sun
Sun
Sun
Sun
none
Sun
Sun
Sun
Sun
LED
DCE
DCM
DCE
DMF
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
DCE
2
4
4
2
2
2
2
2-4
2
2
2
2
2
10, 18a
10, 18a
10, 18a
10, 18a
5, 18a
10, BrCCl3
CCl4
10, CBr4
0.75 h
12 h
12 h
0.5 h
20 h
20 h
20 h
12 h
12 h
15 h
20 h
20 h
1.5 h
79%
71%
[c]
No reaction
10, BrCCl3
5, BrCCl3
10, BrCCl3
5, BrCCl3
10, BrCCl3
10, BrCCl3
10, BrCCl3
10, BrCCl3
10, BrCCl3
10, BrCCl3
60%[d]
[e]
[e]
[e]
5, 18a
10, 18a
10, 18a
None
10,18b
10, 18c
10, 18d
10, 18a
No reaction
9
No reaction
[e]
10
11
12
13
[f]
[f]
65%
[a] 18a=4-dimethylamino-pyridine, 18b=4-methoxy-pyridine, 18c=2,4,6-trimethyl-pyridine, 18d=pyridine. [b] All yields are isolated yields after
purification by flash chromatography. [c] Cosolvent. [d] Partial Fmoc-deprotection was observed. [e] Trace amounts of product were observed. [f] TLC
59 showed complete consumption of amino-acid 16, but no peptide coupling product was observed.
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 8
These are not the final page numbers!

Angewandte
Research Articles
Chemie
1
2
3
4
5
6
7
8
9
Using dry DMF as the solvent (Table 1, Entry 4) led to
(18e) in place of DMAP (1) was carried out (Scheme 3 and
Supporting Information, See Figures S6-S9 for the HRMS
spectra). The reaction was interrupted after one hour and the
mixture was analyzed by HRMS. Since 4-piperidinyl-pyridine
(18e) is a poorer catalyst than DMAP (1) amide formation is
slower, which allowed us to observe, by HRMS, both the
expected product 23 as well as the activated 4-dimethylamino-
pyridinium carboxylate 24.[22] Furthermore, 25 and its hydro-
lyzed congener 26, were observed by HRMS (Scheme 3).
1
2
3
4
5
6
7
8
9
partial hydrolysis of the Fmoc group and thus an isolated 60%
yield of the protected dipeptide. Using 5 equivalents of
DMAP (1/18a) and BrCCl3 (Table 1, Entry 5) or 5 equiva-
lents of either reagent and 10 equivalents of the other
(Table 1, entry 6 and 7) led to incomplete conversion of the
starting material even after extended light irradiation. In
a preliminary experiment the use of blue LED (emission at
395–400 nm) as a light source afforded the protected dipep-
10 tide 20 in 65% isolated yield after 1.5 hours irradiation
11 (Table 1, entry 13, See Figure S4 for a picture of the Blue
12 LED used).
A question arose whether it would be possible to convert 10
DMAP (1) into coupling reagents 7 or 8 using a simple 11
oxidant. However, carrying out the reaction of 2-naphthalene 12
carboxylic acid with hexane amine in the absence bromotri- 13
chloromethane using NBS, with or without light, afforded no 14
discernible amide formation (See Supporting Information). 15
This is likely because chemical oxidants may oxidize the 16
intended coupling partner amine competitively with DMAP. 17
In contrast, only DMAP forms a charge-transfer complex 18
with bromotrichlormethane that absorbs in the visible light 19
region (Figure S1a). It is possible that it is this unique close 20
association that allows DMAP (1) to be converted selectively 21
into coupling reagent 7/8 under irradiation of the charge- 22
transfer complex in the presence of the free amine of the 23
13
Control experiments show that no reaction takes place in
14 the absence of light (Table 1, Entry 8) or in the absence of
15 DMAP (1/18a, Table 1, Entry 9). The sweet spot occupied by
16 the ability of DMAP (1) to function in the described roles that
17 is, in complex formation, as a base, as an acylation catalyst
18 and, in modified form, as the coupling reagent, is accentuated
19 by comparing the reaction using DMAP (1/18a) to those of 4-
20 methoxy-pyridine (18b), 2,4,6-trimethyl-pyridine (18c) and
21 pyridine (18d) under identical conditions (Table 1, Entry 10–
22 12). None of these experiments led to formation of appreci-
23 able amounts of the desired product, even after more than
24 20 hours (in the course of two days) of irradiation.
amino acid coupling partner (Scheme 2a).
24
25
Attempts to isolate the reactive salts 7 or 8 (Scheme 2a)
We next set out to delineate the scope of the method by 25
coupling representative examples of the common amino acids 26
to form dipeptides (Scheme 4). We aimed at testing the 27
stability of a variety of common protecting groups to the 28
reaction conditions. Thus, amino acids protected with the 29
typical amine protecting groups Fmoc, Boc, and CBz groups 30
were tested in the reaction. The coupling reactions of alcohol 31
containing amino acids serine and threonine were performed 32
on protected alcohols using benzyl and tert-butyl ethers, 33
respectively. The phenolic oxygen of tyrosine was also 34
protected as its tert-butyl ether. Aspartic acid, the represen- 35
tative of amino acids with acidic side chains, was protected as 36
its tert-butyl ester in order to differentiate it from the main 37
chain carboxylic acids that were protected as base labile 38
methyl or ethyl esters. All carboxyl protected amino acids in 39
the present study are commercially available as their hydro- 40
chloride salts and were used in this form. Amino acids with 41
basic side chains were protected with either Trityl (histidine) 42
or Boc (tryptophan, proline, and lysine). All the tested 43
protected amino acids were found to be compatible with the 44
reaction conditions. The exception was methionine, presum- 45
ably due to oxidation of the sulfur atom. In contrast, trityl 46
protected cysteine is compatible with the reaction conditions. 47
Asparagine and glutamine were not tested, but as amides they 48
26 formed during one hour of irradiation at 208C proved futile.
27 However, irradiation of a reaction mixture of only DMAP
28 and BrCCl3 at standard concentrations for 36 hours (8 hours
29 of sunlight and 28 hours of white LED (20 W, 6500 K)) led to
30 formation of substantial amounts of compound 8 that could
31 be isolated after concentration and flash chromatography of
32 the reaction mixture. Despite being isolated in an impure
33 form due to its tendency to react with nucleophiles, com-
34 pound 8 could be identified using a combination of 13C and
35 DEPT NMR, and direct injection HRMS (See SI for data and
36 spectra).
37
Further evidence for the mechanism proposed in
38 Scheme 2 was the isolation of oxidized 14 that is, the
39 corresponding formamide 15 in 6% yield in one of our
40 optimization experiments using CCl4 (See SI for structure and
41 characterization data for 15). Unfortunately, this isolation
42 proved difficult to reproduce. Instead, it was attempted to
43 directly detect reaction intermediates using HRMS
44 (Scheme 3). To this end, the reaction of 2-napthoic acid (21)
45 with n-hexane-1-amine (22) using 4-piperidinyl-pyridine
46
47
48
49
should be compatible with the reaction conditions.
49
50
L-phenyl alanine methyl ester was coupled with Fmoc- 50
glycine within 50 minutes and Fmoc-Gly-L-Phe-OMe (31) 51
isolated in 92% yield after standard flash chromatography. 52
Racemic (DL)-phenylalanine was also coupled to Fmoc- 53
glycine to give racemic Fmoc-Gly-DL-Phe-OMe (30) in 90% 54
yield. Importantly, comparison of the racemic (30) and single 55
enantiomer (31) products by chiral HPLC showed that no 56
epimerization took place during reaction, workup or purifi- 57
cation (See Supporting Information). These yields compare 58
well with the 73–83% yield reported in the literature using 59
51
52
53
54
55
56
57
58
59 Scheme 3. HRMS study of the reaction using DMAP analogue 18e.
Angew. Chem. Int. Ed. 2021, 60, 2 – 8
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
1
2
1
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Scheme 4. Scope of the peptide coupling with respect to different amino-acids and protection group schemes.
48 reagents such as DCC (73%), DIC (78%), HBTU (79%) or
integrity was preserved as no other diastereoisomers could be 48
observed. The diastereoisomer of 33 that is, Fmoc-L-Ile-D- 49
Phe-OMe (34) was prepared in 59% yield. HPLC/MS 50
analysis of the diastereoisomers Fmoc-L-Ile-L-Phe-OMe 51
(33) and Fmoc-L-Ile-D-Phe-OMe (34) showed that no 52
detectable epimerization had taken place in either of these 53
49 HCTU (83%).[23]
50
L-Phenylalanine and glycine could also be coupled in the
51 reverse order giving rise to Fmoc-L-Phe-Gly-OEt (32) in
52 82% isolated yield after 45 minutes reaction time. Chiral
53 HPLC again showed that no epimerization took place under
54 the reaction conditions (See SI). Hindered Fmoc-L-isoleucine
55 was coupled to glycine ethyl ester to give 20 in 79% yield, as
56 well as to more hindered L-phenylalanine methyl ester to give
57 Fmoc-L-Ile-L-Phe-OMe (33) in 83% yield within one hour.
58 In this, as well as all reaction involving the coupling of two or
59 more chiral amino acids, NMR indicated that stereochemical
two reactions.
54
The synthesis of Fmoc-L-Ile-L-Phe-OMe (33) was scaled 55
up to give 1.2 gram of product in 83% yield (Scheme 4). This 56
yield is superior to those reported for the same coupling using 57
DCC (72%), DIC (72%), HCTU (77%) and HBTU 58
(71%).[23] In this experiment it was also shown to be possible 59
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 8
These are not the final page numbers!

Angewandte
Research Articles
Chemie
1
2
3
4
5
6
7
8
9
to isolate excess DMAP in 83% yield (92% of the theoretical
product CBz-L-Phe-Gly-OEt (45). CBz protected L-phenyl-
alanine was also coupled with the more hindered L-leucine
methyl ester to afford 46 in 88% yield. In both cases the
reaction was complete within 60 minutes.
It is well known that even di-peptides may have con-
formationally restricted space. This may cause further cou-
pling reactions to become progressively less efficient. To-
wards studying this issue, we prepared three tripeptides
(Scheme 5). The methyl ester of Boc-Gly-L-Phe-OMe (42)
1
2
3
4
5
6
7
8
9
excess of reagent). Recovery of excess bromotrichlorome-
thane was attempted using simple short path distillation in
lieu of concentration of the reaction mixture on a rotary
evaporator. This allowed its recovery in 29% yield.
Fmoc-(O-tert-butyl)-L-threonine was coupled with gly-
cine ethyl ester to afford Fmoc-L-Thr(tBu)-Gly-OEt (35) in
85% yield. The tert-butyl ester and Fmoc protected aspartic
acid was coupled with glycine ethyl ester to produce Fmoc-L-
10 Asp(OtBu)-Gly-OEt (36) in 89% isolated yield. The phenolic
11 amino acid tyrosine in its O-tert-butyl and N-Fmoc protected
12 form was coupled with L-phenyl-alanine methyl ester giving
13 Fmoc-L-Tyr(tBu)-L-Phe-OMe (37) in 78% isolated yield. L-
14 Cysteine protected at its sulfur atom by trityl and at its
15 nitrogen by Fmoc was successfully coupled with glycine ethyl
16 ester leading to isolation of Fmoc-L-Cys(Trt)-Gly-OEt (38) in
17 77% yield with no indication of oxidation of sulfur. L-
18 Histidine, likewise, protected with trityl at its imidazole
19 nitrogen and Fmoc at its main chain amine function was
20 coupled with glycine ethyl ester to give Fmoc-L-His(Trt)-Gly-
21 OEt (39) in 71% yield. Differentially protected L-lysine, that
22 is, with Boc protected side chain amine and Fmoc on the main
23 chain amine, was successfully linked to L-phenylalanine
24 methyl ester giving Fmoc-L-Lys(Boc)-Leu-OMe (40) in
25 85% yield within 80 minutes. Finally, the complex amino
26 acid L-tryptophan with its indole nitrogen protected with Boc
27 and its basic nitrogen protected with Fmoc was coupled with
28 L-phenylalanine methyl ester to afford Fmoc-L-Trp(Boc)-L-
29 Phe-OMe (41) in 78% yield (Scheme 4). For comparison
30 Fmoc-L-Trp(Boc)-L-Phe-OMe (41) has been reported to be
31 prepared in 52% yield using O-(1H-6-Chlorobenzotriazole-1-
32 yl)-1,1,3,3-Tetramethyl Uronium hexafluoro phosphate
33 (HCTU)/DiIsoPropylEthylAmine (DIPEA)[24] and in 61%
34 yield using HBTU/DIPEA[25] (12 h reaction time in both
35 cases). There was no oxidation of the aromatic side chains in
36 neither tyrosine nor tryptophan under our reaction condi-
37 tions.
was hydrolyzed under basic conditions to give the free acid 49 10
in 84% yield after purification. Subsequently it was coupled 11
to L-leucine methyl ester (2 equiv) to give the protected 12
immunomodulating tripeptide Boc-Gly-L-Phe-L-Leu-OMe 13
(50)[28] in 86% yield after one hour of reaction time and 14
purification via flash chromatography on a 179 mg scale. 15
NMR showed no discernible epimerization compromising the 16
two chiral centers in the molecule.
17
The ester of Boc-(O-Bzl)-Ser-L-Leu-OMe (47) was hy- 18
drolyzed as described above in 83% yield and coupled with L- 19
alanine methyl ester (Scheme 5). This provided the purported 20
deliciously tasting peptide L-Ser-L-Leu-L-Ala (52)[28] in its 21
protected form in 89% yield. None of the three chiral centers 22
were epimerized to any detectable degree during the reaction 23
as evidenced by NMR. In a third example, Fmoc-L-Ile-L-Phe- 24
OMe (33) was hydrolyzed with lithium hydroxide and the free 25
acid 53 isolated in 84% yield. Coupling of this hindered 26
dipeptide to Glycine ethyl ester provided the corresponding 27
protected tripeptide Fmoc-L-Ile-L-Phe-Gly-Oet (54) in 72% 28
yield.
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
38
In examining Boc protected amino acids, simple N-Boc-
39 glycine was coupled with L-phenylalanine methyl ester to give
40 Boc-Gly-L-Phe-OEt (42) in 87% isolated yield. This is
41 comparable to the 92% yield achieved in forming 31 with
42 Fmoc protected glycine described above. N-Boc-L-proline
43 was coupled with glycine ethyl ester to give Boc-L-Pro-Gly-
44 OEt (43) in 89% yield within 60 minutes. This is quite
45 comparable to the 95% yield reported after overnight
46 reaction with HOBt/EDC/DIPEA.[26] Boc protected L-leu-
47 cine was coupled to the hindered secondary nitrogen of L-
48 proline methyl ester. After 180 minutes the protected dipep-
49 tide Boc-L-Leu-L-Pro-OMe (44) was isolated in 63% yield.
Scheme 5. Synthesis of tripeptides. [a] LiOH (3 equiv), THF:H2O (4:1),
08C, 40 min. [b] Sunlight (>330 nm), specified amino acid hydrochlo-
ride (2 equiv), DMAP (10 equiv), BrCCl3 (10 equiv) DCE, 1 h.
50 This compound has previously been prepared by Gawley Conclusion
51 using HOBt/EDC/Et3N in 88% yield after ten hours reaction
52 time.[27] O-Benzyl protected N-Boc-L-serine was successfully
53 coupled with both unhindered glycine ethyl ester and
54 hindered L-leucine methyl ester to give 48 in 86% and 47 in
55 91% yield, respectively after 60–65 minutes reaction. Benzyl
56 ethers are known to be susceptible to oxidation, yet no sign of
57 debenzylation was observed.
In summary, we have developed a solar light-assisted 52
peptide coupling reaction based on in situ formation of 53
a novel peptide coupling principle (7/8) derived from DMAP 54
(Scheme 2). The reaction proceeds in a smooth manner, 55
under mild conditions without epimerization, and in yields 56
comparable to those obtained using state of the art coupling 57
reagents. However, in contrast to standard peptide coupling 58
methods no additional reagents, inert atmosphere or pre- 59
58
We studied CBz protected L-phenylalanine in the reaction
59 with glycine ethyl ester resulting in a 91% isolated yield of the
Angew. Chem. Int. Ed. 2021, 60, 2 – 8
ꢀ 2021 Wiley-VCH GmbH
www.angewandte.org
&&&&
These are not the final page numbers!

Angewandte
Research Articles
Chemie
3038; e) D. Staveness, I. Bosque, C. R. J. Stephenson, Acc.
Chem. Res. 2016, 49, 2295 – 2306.
[10] a) N. Kannan, D. Vakeesan, Renewable Sustainable Energy Rev.
2016, 62, 1092 – 1105; b) A. Kleidon, L. Miller, F. Gans, in Top.
Curr. Chem., Vol. 371 (Eds.: H. Tꢁysꢁz, C. K. Chan), Springer,
Berlin, 2016.
[11] D. M. Schultz, T. P. Yoon, Science 2014, 343, 1239176 – 1239176.
[12] M. Oelgemçller, Chem. Rev. 2016, 116, 9664 – 9682.
[13] a) G. N. Papadopoulos, C. G. Kokotos, J. Org. Chem. 2016, 81,
7023 – 7028; b) N. Iqbal, E. J. Cho, J. Org. Chem. 2016, 81, 1905 –
1911; c) G. Pandey, S. Koley, R. Talukdar, P. K. Sahani, Org. Lett.
2018, 20, 5861 – 5865; d) S. Gaspa, I. Raposo, L. Pereira, G.
Mulas, P. C. Ricci, A. Porcheddu, L. De Luca, New J. Chem.
2019, 43, 10711 – 10715.
[14] a) S. Das, S. Ray, A. B. Ghosh, P. K. Samanta, S. Samanta, B.
Adhikary, P. Biswas, Appl. Organomet. Chem. 2018, 32, e4199;
b) H. Liu, L. Zhao, Y. Yuan, Z. Xu, K. Chen, S. Qiu, H. Tan, ACS
Catal. 2016, 6, 1732 – 1736; c) W. Song, K. Dong, M. Li, Org. Lett.
2020, 22, 371 – 375.
[15] I. Cohen, A. K. Mishra, G. Parvari, R. Edrei, M. Dantus, Y.
Eichen, A. M. Szpilman, Chem. Commun. 2017, 53, 10128 –
10131.
[16] Y. Zhang, D. Riemer, W. Schilling, J. Kollmann, S. Das, ACS
Catal. 2018, 8, 6659 – 6664.
[17] a) W. J. Lautenberger, E. N. Jones, J. G. Miller, J. Am. Chem.
Soc. 1968, 90, 1110 – 1115; b) A. K. Mishra, G. Parvari, I. Cohen,
N. Fridman, Y. Eichen, A. M. Szpilman, J. Coord. Chem. 2018,
71, 2082 – 2089.
[18] W. Steglich, G. Hçfle, Angew. Chem. Int. Ed. Engl. 1969, 8, 981;
Angew. Chem. 1969, 81, 1001.
[19] pKa is often used as an indicator for leaving group ability.
Glucose and other hexopyranoses generally have a pKa of app.
12 and this is ascribed to the electron withdrawing effect of the
hemiacetal ring oxygen (See: J. A. Rendleman, in Carbohydrates
in Solution, Vol. 117, ACS, Washington, 1973, pp. 51 – 69.) A
similar effect should be expected for hemiaminals making
hemiaminal esters activated in comparison with esters of simple
alcohols (pKa 15 – 18).
[20] Y. Zhao, D. G. Truhlar, Theor. Chem. Acc. 2008, 120, 215 – 241.
[21] F. L. Hirshfeld, Theor. Chim. Acta 1977, 44, 129 – 138.
[22] V. Lutz, J. Glatthaar, C. Wꢁrtele, M. Serafin, H. Hausmann, P. R.
Schreiner, Chem. Eur. J. 2009, 15, 8548 – 8557.
[23] V. Erapalapati, U. A. Hale, N. Madhavan, Tetrahedron Lett.
2019, 60, 151311.
[24] B.-T. Xin, G. de Bruin, E. M. Huber, A. Besse, B. I. Florea, D. V.
Filippov, G. A. van der Marel, A. F. Kisselev, M. van der Stelt, C.
Driessen, M. Groll, H. S. Overkleeft, J. Med. Chem. 2016, 59,
7177 – 7187.
[25] M. J. Terrey, A. Holmes, C. C. Perry, W. B. Cross, Org. Lett. 2019,
21, 7902 – 7907.
[26] Y. Wei, J. Li, Z. Chen, F. Wang, W. Huang, Z. Hong, J. Lin, Eur. J.
Med. Chem. 2015, 101, 409 – 418.
[27] T. K. Beng, R. E. Gawley, J. Am. Chem. Soc. 2010, 132, 12216 –
12217.
[28] I. Gill, R. Lꢂpez-FandiÇo, X. Jorba, E. N. Vulfson, Enzyme
Microb. Technol. 1996, 18, 162 – 183.
1
2
3
4
5
6
7
8
9
dried solvents are necessary. In general, reaction times of
about 1 hour are required. The method is instantly applicable
by others as it is compatible with standard amino acid
protection groups and most common amino acids. Excess
DMAP may be recycled. The method has been scaled to
prepare dipeptide on gram scale and has potential for further
scale-up using existing flow technology.[29]
1
2
3
4
5
6
7
Future work will focus on the development of a solid-
phase compatible with light-assisted peptide synthesis. The
8
9
10 use of for example, a laser as the light source, would pave the
11 way for many other interesting applications, such as high
12 density, spatially localized synthesis of protein sequences for
13 functional devices such as protein chips, as well as for light
14 induced 3D-printing of polyamides. Extensions in these
15 directions will be reported from our laboratories in due
16 course.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
17
18
19
20
Acknowledgements
21 This research was supported in part by an Israel Science
22 Foundation Personal Grant (Grant no. 1914/15) to A.M.S. Dr.
23 Mark A. Iron is gratefully acknowledged for helpful com-
24 ments regarding the calculation aspects of this paper. Dr. Guy
25 Kamnesky is gratefully acknowledged for assistance in
26 preliminary epimerization analyses.
27
28
29
30
Conflict of interest
31 The authors declare no conflict of interest.
32
33 Keywords: charge-transfer complex · halogen bonding ·
34 peptide coupling · solar light · visible light
35
36
[1] T. Uhlig, T. Kyprianou, F. G. Martinelli, C. A. Oppici, D.
37
Heiligers, D. Hills, X. R. Calvo, P. Verhaert, EuPa Open
Proteomics 2014, 4, 58 – 69.
[2] K. J. Jensen, P. T. Shelton, S. L. Pedersen, Peptide Synthesis and
Applications, Springer, Berlin, 2013.
[3] V. R. Pattabiraman, J. W. Bode, Nature 2011, 480, 471 – 479.
[4] A. A. Zompra, A. S. Galanis, O. Werbitzky, F. Albericio, Future
38
39
40
41
42
Med. Chem. 2009, 1, 361 – 377.
[5] a) N. Hartrampf, A. Saebi, M. Poskus, Z. P. Gates, A. J. Callahan,
43
44
A. E. Cowfer, S. Hanna, S. Antilla, C. K. Schissel, A. J.
Quartararo, X. Ye, A. J. Mijalis, M. D. Simon, A. Loas, S. Liu,
C. Jessen, T. E. Nielsen, B. L. Pentelute, Science 2020, 368, 980 –
987; b) B. L. Bray, Nat. Rev. Drug Discovery 2003, 2, 587 – 593.
45
46
47
[6] J. C. Sheehan, G. P. Hess, J. Am. Chem. Soc. 1955, 77, 1067 –
48
1068.
49
[7] A. El-Faham, F. Albericio, Chem. Rev. 2011, 111, 6557 – 6602.
[8] a) B. Trost, Science 1991, 254, 1471 – 1477; b) P. A. Wender,
Tetrahedron 2013, 69, 7529 – 7550; c) T. Newhouse, P. S. Baran,
R. W. Hoffmann, Chem. Soc. Rev. 2009, 38, 3010 – 3021.
[9] a) C. K. Prier, D. A. Rankic, D. W. C. MacMillan, Chem. Rev.
[29] a) Z. J. Garlets, J. D. Nguyen, C. R. J. Stephenson, Isr. J. Chem.
2014, 54, 351 – 360; b) D. Cambiꢃ, C. Bottecchia, N. J. W.
Straathof, V. Hessel, T. Noꢄl, Chem. Rev. 2016, 116, 10276 –
10341.
50
51
52
53
2013, 113, 5322 – 5363; b) M. H. Shaw, J. Twilton, D. W. C.
MacMillan, J. Org. Chem. 2016, 81, 6898 – 6926; c) N. A.
Romero, D. A. Nicewicz, Chem. Rev. 2016, 116, 10075 – 10166;
d) X. Lang, J. Zhao, X. Chen, Chem. Soc. Rev. 2016, 45, 3026 –
54
Manuscript received: August 22, 2020
Accepted manuscript online: February 23, 2021
55
56
Version of record online: && &&
,
&&&&
57
58
59
&&&&
www.angewandte.org
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 2 – 8
These are not the final page numbers!