Highly Productive Continuous Flow Synthesis of Di- and Tripeptides in Water

Article abstract of DOI:10.1021/acs.oprd.7b00214

The reaction of amino acid derived N-carboxyanhydrides (NCAs) with unprotected amino acids under carefully controlled aqueous continuous flow conditions realized the formation of a range of di- and tripeptide products in 60-85% conversion at productivities of up to 535 g·L^-1h^-1. This required a fundamental understanding of the physicochemical aspects of the reaction resulting in the design of a custom-made continuous stirred tank reactor (CSTR) with continuous solids addition, high shear mixing, automated pH control to avoid the use of buffer, and efficient heat removal to control the reaction at 1 ± 1 °C.

Full text of DOI:10.1021/acs.oprd.7b00214

Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES
Full Paper
Highly Productive Continuous Flow Synthesis of Di- and Tripeptides in Water
Katherine E Jolley, William Nye, Carlos González Niño,
Nikil Kapur, Alain Rabion, Kai Rossen, and A. John Blacker
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.7b00214 • Publication Date (Web): 30 Aug 2017
Downloaded from http://pubs.acs.org on August 30, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Organic Process Research & Development is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Highly Productive Continuous Flow Synthesis of
Di- and Tripeptides in Water
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
60
Katherine. E. Jolley, † William. Nye,† Carlos González Niño,† Nikil Kapur,†
Alain Rabion,§ Kai Rossen ‡ and A. John Blacker †*
† Institute of Process Research and Development. School of Chemistry and School of Chemical
and Process Engineering and School of Mechanical Engineering, University of Leeds,
Woodhouse Lane, LS2 9JT, West Yorkshire, UK.
§ Sanofi-Aventis France, LGCR-CBD Chemical Development, Toulouse, France
‡ Sanofi-Aventis Deutschland GmbH, R&D LGCR / Chemical Development FF Industriepark
Hoechst, Bldg. G838, Room 010, D-65926 Frankfurt am Main, Germany
* Corresponding authors. E-mail: j.blacker@leeds.ac.uk;
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 27
1
2
3
4
ABSTRACT
5
6
7
8
9
The reaction of amino acid derived N-carboxyanhydrides (NCAs) with unprotected amino acids
under carefully controlled aqueous continuous flow conditions realized the formation of range of
di- and tripeptide products in 60-85% conversion at productivities of up to 535 g.L-1h-1. This
required a fundamental understanding of the physicochemical aspects of the reaction resulting in
the design of a bespoke continuous stirred tank reactor (CSTR) with continuous solids addition,
high shear mixing, automated pH control to avoid the use of buffer, and efficient heat removal to
control the reaction at 1±1 °C.
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
60
KEYWORDS
Small Peptide; Leuch’s NCA; Water; Continuous flow CSTR; Process control
INTRODUCTION
Short peptides, that are composed of several amino acids, are important compounds within society
with applications in medicine,1 farming,2 food,3 and materials,4 however their manufacture is both
inefficient and unproductive.5, 6 Merrifield’s solid-phase-peptide-synthesis (SPPS) is remarkably
predictable and used widely for making small quantities of complex peptides.7 However, the scale-
up of SPPS to make even kilograms of short peptides is difficult, being characterised by:
sequences of excess coupling reagents; protection-deprotection steps; dilute conditions with
problematic organic solvents; addition and deletion sequences caused by physical issues with
washing the resin, Scheme 1a and Table 2.8, 9
ACS Paragon Plus Environment
2

Page 3 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
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
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Examples of dipeptide synthesis indicating atom economy in colored type, with waste
and operations in black type. a. SPPS;9 b. SPS;10 c. Enzyme.11
On the other hand, organic solution phase synthesis (SPS), that uses similar chemical methodology
to SPPS, is the main method for small peptide manufacture. It also suffers poor green credentials:
protecting groups, atom inefficiency, high E-factor and low process mass intensity, Scheme 1b,
Table 2.12-14 Biosynthesis methods are leading the way in synthesis of longer peptides, for example
in insulin manufacture,15 but are unproductive for short peptides. For example proteases are used
to couple protected amino acids, but rely on high enzyme loadings and displacing unfavourable
equilibria with either excess reagents, solubility effects or organic solvents, Scheme 1c.11, 16-22
Surprisingly, the pioneering work done in the 1950-60s by Bartlett and Hirschmann describing the
coupling of N-carboxyanhydride (NCA or Leuch’s anhydride) to amino acids has not been widely
adopted, despite it being productive, cost effective and green, Scheme 2.23-27
Scheme 2. The NCA method of peptide synthesis
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 27
1
2
3
4
5
6
7
8
9
All twenty amino acids have been tested with several NCAs, though some require side-chain
protection, and 4g of a pentapeptide has been prepared in overall 27% yield, though the purity was
not measured directly.24 A key enabler in this methodology is the use of water, which unlike
organic solvents used in SPS is able to dissolve high concentrations of the free amino acids. In this
article, we report studies to overcome the physico-chemical problems with the NCA methodology
that appear to be a barrier to its exploitation, and initial development of a continuous process that
enables precise control of the conditions and which required an unusual reactor design. This was
used to produce a variety of di- and tripeptides in good conversion and high productivity, including
an Aspartame derivative.28
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
60
Hirschmann’s NCA process involves the batch reaction of -amino carboxylates with solid NCAs
under carefully determined aqueous conditions: pH 10.2, controlled using a 1M sodium borate
buffer with 6M KOH addition; a temperature of 0 ᵒC maintained with ice.27 Control of these are
important, otherwise side reactions occur that are exacerbated by the rapid formation of acid and
a strong exotherm. Our studies found that at this temperature the sodium borate buffer precipitates,
furthermore, addition of ice dilutes the reaction and causes problems with mixing. Accordingly,
less buffer and spatially separate cooling are preferable. It has been observed that if the NCA is
dissolved in a dry organic solvent and added to a solution of the amino acid, low yields occur as a
result of side reactions.29 In this regard, Hirschmann recognized the need for high shear mixing to
rapidly disperse the highly reactive solid NCA which, when added portion-wise into the best
available equipment, a “Waring” kitchen blender, improved the selectivity.25 High shear mixing is
used to reduce the depth of the static boundary layer on a particle, and is a function of the Reynold’s
number, in particular the fluid velocity, particle shape and size, Figure 1.30
ACS Paragon Plus Environment
4

Page 5 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
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
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. A model of the reactions occurring with NCA in the liquid boundary layer and bulk
solution during dissolution of the solid.
The amino acid needs to be brought rapidly into contact with the NCA to compete with its
hydrolysis, also to neutralise the acid to prevent decarboxylation. Any CO2 that is generated is
trapped by the base as carbonate. The reason that pH10.2 has been found to be optimal becomes
clear: the NCA pKa is 11.3, so this pH keeps the reactive anion concentration low; whilst the
amino acid pKa is 9.2, keeping the concentration of the free amine high. The reactivity of NCAs
has been discussed in detail by Kricheldorf, and includes their use in the formation of
homopolypeptides.31 Bartlett studied the kinetics of NCA hydrolysis and estimated the rate to be
one third that of the amino acid at the same pH, both rates being above 1.6*106 M-1.s-1.32, 33 Caplow
and Johnson each showed the intermediate carbamic acid is unstable above 0 ᵒC (a problem in
water) and can prematurely decarboxylate to the free amino acid with a rate constant of >9* 107
M-1.s-1 for glycine at pH10,34, 35 leading to an unwanted tripeptide, Figure 1. Hirschmann also
proposed the formation of hydantoic acid side-products by decomposition of the NCA to
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 27
1
2
3
4
5
6
7
8
9
isocyanate, as well as NCA dimerization.26 These side reactions lead to impurities that are difficult
to separate from the desired product, Figure 1, and it is clear that carrying out the reaction in batch
makes control of the conditions difficult as others have observed.26 Previous reports of solution
phase peptide formation in flow include Knudsen’s automated sequence specific synthesis where
Cbz protected NCAs formed in-situ are reacted with unprotected amino acids in DMF.36 Whilst
achieving excellent yields for a range of peptides, the process is of low productivity, requires
protecting groups and excess NCA that reduce the atom efficiency. Fuse et al. report the reaction
of O-protected amino acids with N-protected amino acids activated by in-situ reaction with
triphosgene,37 high yields are achieved with a 5 second residence time (Tres) over 2 steps, but with
similar limitations in efficiency. Herein we describe initial studies to develop a continuous-mode
version of Hirschmann’s NCA process aimed at overcoming these limitations.
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
60
RESULTS AND DISCUSSION
Our study selected a single-stage CSTR in preference to a plug-flow reactor (PFR) which is
incapable of providing solids addition, high intensity multiphase mixing, and pH control, Figure
2. The size and shape of the reactor were defined by the equipment needed to operate the process.
A problem with the single stage CSTR is that the residence time distribution (RTD) can be wide
with material passing straight through or never emerging, use of a food dye marker indicated more
than 4 reactor volumes were required. A multi-stage CSTR would give a more efficient process,
however with the process limitations the complexity of such a system precluded this approach.
The purpose-built reactor comprised: a stainless steel vessel with overflow tube to maintain 300
mL volume; an automated powder dispenser to continuously dose the NCA (achieved with a
variable speed, motor-driven, rotating angled pipe, accurate to 7 wt%, (SI, 7.2)); a high shear mixer
ACS Paragon Plus Environment
6

Page 7 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
employed to promote rapid mixing of the reactants; an internal cooling-coil required to maintain
the desired temperature of 0-2 ᵒC; pH probe and thermocouple. The heat from the mixer at 4000
rpm along with 1.8 mL/min KOH caused the water temperature to rise at 0.85 ᵒC.min-1; and by
difference, the reaction exotherm was found to be 0.54 ᵒC.min-1, 9.8 kJ.mol.-1min-1 (SI, Figure 1).
In initial experiments, the amino acid was dissolved in sodium borate buffer at pH 10.2 and
pumped at a rate stoichiometric with the continuously added, solid NCA,38 which defined the
residence time by the overflow pipe.
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
60
Figure 2. Diagram and photographs of the CSTR set-up used, showing high shear mixer,
continuous solids addition system, internal cooling coil, and pumps.
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 27
1
2
3
4
5
6
7
8
9
The actual residence time was determined using a pulse of dye and found to be 8.6 minutes
compared to 7 minutes theory, indicating only a small hold-up (SI 2.0). To achieve steady-state 2-
3 reactor volumes (one RV is 300 mL) were required, and residence times of 5, 7 and 14 minutes
were evaluated to determine the optimum (SI 2.0); equating to 10-70g of NCA per experiment.
Initial experiments employed commercially available L-Ala NCA which was reacted with L-Phe,
to form L-Ala-L-Phe dipeptide.39 Since litres of aqueous product were generated in each
experiment, the amino acid to dipeptide conversion was measured by sampling the acid quenched
solutions, and determining the amino acid and peptide concentrations against authentic standards
by 1HNMR, HPLC and LCMS. The instability of the NCA meant that its conversion to dipeptide
could not be measured directly. Figure 3 shows the importance of mixing, measured as mixer speed
vs. conversion.
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
60
Figure 3. Reaction conditions and the effect of mixer speed on dipeptide formation
Using a residence timeof 7 min, 3*300 mL reactor volumes were collected at each speed. At 1200
rpm product formation was low and inconsistent with the solid NCA building-up on the surface
ACS Paragon Plus Environment
8

Page 9 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(ESI Figure 4). Above 3000 rpm, 70% of the crude dipeptide was produced with little variation at
steady-state, and 4000 rpm was selected for subsequent reactions. The reaction stoichiometry is
important since it was found that 1.5 eq. of NCA led to over-reaction, with tripeptide and other
side-products formed, whilst an undercharge caused problems with amino acid and dipeptide
separation; all subsequent experiments used 1.1 eq. of NCA (ESI Figure 5). Table 1 shows the
results of varying other reaction parameters.
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
60
Table 1. Optimization of reaction conditions for L-Ala-L-Phe formation.
Entry[a]
[Na2B4O7]aq. (M)[b] Achieved
pH[c]
Achieved
temp. (ᵒC)
% Dipeptide in
crude reaction
product [d]
1
0.4
9.9-10.3
10-10.35
8.8-9.2[e]
10.8-11.3[f]
9.6-10.3
9.5-10.5
9.9-10.5
10-10.6[i]
10.2-10.4[j]
10.1-10.2
9.8-10.2
9.9-10.4
-1.3-2.7
1.7-3.1
0.8-2.2
0.0-3.3
0.5-4.0
0.6-1.7
4.2-6.4
2.0-4.6
0.5-3.7
3.3-5.6
0.9-1.5
1.0-2.1
51
63
51
70
72
68
39
58
33
61
67
71
2
0.1
3
0.1
4
0.1
5[g]
6
0.1
None
0.1
7[h]
8
None
None
0.1
9
10[g][k]
11[l]
12[k][l]
0.1
0.1
[a] Tres = 7 min. [b] With 0.2M amino acid initially at pH 10.2. [c] Manual addition of 6M KOH
1
to target pH 10.2 [d] Calculated from the HNMR ratio of dipeptide : amino acid + dipeptide +
tripeptide.[e] target pH 9 [f] target pH 11. [g] Tres = 5 min. [h] NCA added as 0.22 M solution in
THF. [i] Et3N used to control pH. [j] NH4OH used to control pH. [k] L-Ala NCA reacted with L-
Leu. [l] Tres = 14 min.
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 27
1
2
3
4
5
6
7
8
9
Whilst 0.4 M sodium borate buffer solution enabled control of pH 10.2, its poor solubility at 0-2
ᵒC lead to difficulties in maintaining constant addition (Entry 1). A lower buffer strength (Entry
2), or no buffer (Entry 6), gave an increase in dipeptide; however relying on manual KOH addition
made it difficult to maintain a constant pH, Figure 4.
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
60
Figure 4. pH control of dipeptide formation with and without pH controller. With manual pH
control the KOH flow rate was adjusted by hand to maintain pH 10.2.
Table 1 Entry 3 shows that at pH 8.8-9.2, 20% less product is produced, whilst a higher pH of up
to 11.3, was less problematic (Entry 4). Increasing the amino acid feed-rate, to reduce the residence
time from 7 to 5 minutes, resulted in poor temperature control, reaching 4 ᵒC (Entry 5) and 5.6 ᵒC
(Entry 10) because of the increased energy required to cool the ambient temperature amino acid
feed. Increasing the Tres from 5 to 7 to 14 minutes, at similar pH and temperature ranges, changed
little the steady-state conversions of 70%, 72%, 67%, (Entries 4, 5, 11) indicating the reactions
were complete even at the shortest time. On the other hand, the reaction of L-Ala NCA and L-Leu,
with residence times of 5 and 14 minutes, improved the conversion from 61% to 71%, that might
ACS Paragon Plus Environment
10

Page 11 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
indicate a slower reaction or reduced mass transfer (Entries 10 and 12). The addition of L-Ala
NCA dissolved in THF gave a significantly lower yield, 39% compared to 72% with solid addition,
confirming previous studies.27, 29 The use of aqueous triethylamine or ammonia instead of KOH,
gave lower yields with reduced thermal control and, in the case of the latter, a major new side-
product, alanine amide (Entries 8 and 9).40 It is interesting to note that the overall quantity of
hydroxide required to maintain the reaction at pH10.2 was found to be 1.61 equivalents relative to
the amino acid (See ESI, Section 1). The titration stoichiometry can be rationalised by considering
the 72% conversion of L-Phe and L-Ala NCA to L-Phe-L-Ala (Entry 5), that consumes 0.72
equivalents of base. Since 1.15 equivalents of NCA are added, this means 0.43 equivalents of base
are used in hydrolysis, and the remaining 0.46 probably react with the CO2 resulting from
carbamate decarboxylation, to form a 1:2 mixture of potassium bicarbonate and carbonate giving
it a pKa of 10.6 in this salt solution compared with the literature pKa of 10.3 in pure water.41 Based
on this, 63% of the NCA forms dipeptide product and 37% is hydrolysed. The close correlation
between calculated and theory stoichiometry confirms Bartlett’s observation that the relative rate
of peptide formation is twice that of the NCA hydrolysis, but also indicates that the side reactions
are difficult to suppress.
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
60
Isolation of small quantities of the products analysed by HPLC and 1HNMR showed no evidence
of racemisation, and this is supported by reports in the literature.25, 37 For purification and isolation,
Hirshmann et al. report the use of chromatography with ion exchange resins, silica gel or carbon
as the stationary phase, depending on the peptide. In our hands, ion exchange chromatography was
successful in separating the salts from the reaction solution, but did not usefully separate the
unreacted amino acid from the desired dipeptide. For reactions carried out without buffer, silica
gel chromatographic purification was possible and gave the desired dipeptide cleanly for L-Ala-
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 27
1
2
3
4
5
6
7
8
9
L-Phe, and in enhanced purity for L-Val-L-Phe. All products, whether prepared with or without
the presence of the buffer, could also be isolated cleanly on a small scale by use of mass directed
preparative chromatography. Tripeptides could not be isolated in this way.
The defined process conditions and controls were then used to further explore the reaction scope,
Figure 5.
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
60
1
[a]
Conversions calculated from the HNMR ratios Dipeptide starting material was 1.3 M crude
[b]
solution from the L-Ala-L-Phe reaction.
Conversion given for dipeptide to tripeptide
transformation. [c] Dipeptide starting material was 0.17 M solution
Figure 5. Peptide products synthesized, molar conversions (product / amino acid) and
productivities.
L-Ala and L-Val NCAs were reacted with L-Phe and L-Leu amino acids and gave 71-73%
conversions, with productivities up to 330 g.L-1.h-1 and unisolated dipeptide masses of 43-58g
which is remarkable when compared with either SPPS or organic phase SPS. The dipeptides were
then used directly as 1.3 M and 0.17 M pH 10.2 solutions and fed into the next reaction with L-
Ala and L-Val NCA respectively. The L-Ala-L-Ala-L-Phe tripeptide was produced in this step
ACS Paragon Plus Environment
12

Page 13 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
with 60% conversion and at 205 g.L-1.h-1 which over 5RV equates to 36g, whilst L-Val-L-Ala-L-
Leu was formed in 85% conversion at 285 g.L-1.h-1, giving 50g of the tripeptide in solution, Figure
5.
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
60
These results show that the conversions and productivities depend upon the type of NCA and
amino acids and further improvement might be achieved by optimizing each reaction separately.
In the meantime, it was decided to make a more interesting tri-peptide and we chose to make an
Aspartame derivative as the L-Asp-L-Phe-OMe dipeptide ester is inexpensive and commercially
available in large quantities. After hydrolysing the methyl ester to give L-Asp-L-Phe it was fed
continuously into the CSTR along with L-Val NCA. With good control of pH and temperature and
1
based on HNMR integration of a sample, the tripeptide was produced at steady-state in 66%
conversion and at a rate of 535 g.L-1.h-1, providing over 5RV 94g. The 1NMR spectrum showed
29% L-Asp-L-Phe starting material and therefore only 5% other impurities.
Presently, the difficulty in preparing longer peptides is the need to separate them from large
quantities of aqueous salt solutions. The small-scale purifications using ion exchange and silica
demonstrate the potential for larger scale; and since the main separation is binary (i.e. amino acid
from dipeptide), continuous purification by simulated moving bed chromatography may be an
option. Further work is looking at the use of membranes and affinity tags. The system is currently
limited by the need for solid addition, the size of mixer and of the reactor, miniaturisation of this
would allow a shorter residence time and fewer side-reactions. If quantitative conversion of
reagents to the desired product could be achieved this would negate the need for intermediate
purifications.
The sodium borate buffer not only poses health risks but is a significant part of the process waste.
An automated pH controller was constructed in-house to allow rapid measurement and adjustment
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 27
1
2
3
4
5
6
7
8
9
of pH. The unit is a proportional integral controller built on an embedded electronic circuit and
tuned for this application (see ESI). The controller is connected to a pH probe within the reaction
solution and to a peristaltic pump with 6M KOH (aq). Figure 4 shows a comparison of manual and
automated pH control for formation of a dipeptide without the use of sodium borate buffer. For
manual pH control, the pH varies by 1 pH unit during the reaction, whereas with automated pH
control the variation is 0.2 pH units. The conversion to product is comparable for both methods,
however the pH controller allows more efficient use of KOH. The continuous process, with
automated pH control allowing the removal of the buffer from the reaction not only offers high
productivity and efficiency, but also avoids safety considerations and significantly reduces the
chemical mass intensity of the process.
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
60
A particular focus of the study was to develop a method with improved environmental
performance, so metrics of related continuous and batch processes were compared, Table 2.42
ACS Paragon Plus Environment
14

Page 15 of 27
Organic Process Research & Development
1
2
3
4
Table 2. Comparison of reaction metrics for formation of dipeptide. [a]
5
6
7
8
Entry
Metric
SPPS
NCA method
(batch)[b]
This study:
un-buffered
(continuous)
L-Val-L-Phe
82[d]
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
48
49
50
51
52
53
54
55
56
57
58
59
60
Peptide
L-Leu-L-Ala[c]
L-Ala-L-Phe
1
2
3
4
5
6
Yield (%)
STY (g.L-1h-1)
Total PMI
RME (%)
AE (%)
50
70
0.014
0.35
686
57
60.5
3165
28
4
67
10
84
86
Health & Safety
DMF
Na2B4O7
None
[a]
[b]
[c]
for calculations see ESI
based on reference 26
estimated from reported tetrapeptide
synthesis reference 9 [d] by HPLC.
Since there is no reported procedure for making the identical peptides, we based it as fairly as
possible on Merrifield’s SPPS description of one stage in making a tetrapeptide, and that of
Hirshmann’s L-Phe-L-Ala in batch.9 Comparing the SPPS and flow NCA method the latter gave
a higher steady-state conversion of 82%, (Entry 1) and was 4600 times more productive (Entry 2),
with the total mass intensity which includes solvent, a hundred-fold better (Entry 3). Reaction mass
efficiency is the reciprocal of E-factor and considers atom economy, yield and stoichiometry, and
this is 67% compared to 4% for the SPPS process (Entry 4). Unsurprisingly without protecting
groups, the batch and flow NCA methods have much superior atom economies than SPPS (Entry
5). Meanwhile comparing the batch and flow NCA processes, the latter is 173-times more
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 27
1
2
3
4
5
6
productive (Entry 1), and has a total mass intensity 24-times better (Entry 2), mainly due to the
avoidance of borate buffer which also improves health-safety aspects (Entry 6).
7
8
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
48
49
50
51
52
53
54
55
56
57
58
59
60
CONCLUSIONS
A continuous flow stirred tank reactor has been carefully designed to control within a narrow
range, operating conditions that favour the reaction of NCA with amino acid, and minimise side
reactions that produce impurities that are difficult to separate. The use of a cooling coil inside the
reactor removes both the reaction exotherm and heat generated from the mixer to maintain the
optimal of 0-2oC. This temperature range is important in lowering the rate of side-reactions relative
to the desired reaction. The need for intense, high shear mixing has been shown to improve amino
acid conversion to the product. This facilitates the rapid dispersion of solid NCA and concentrated
hydroxide that is added to neutralise the acid produced in the reaction, that otherwise leads to
product decarboxylation. Since the reactions in the system are very fast, they are likely to take
place near the surface of the NCA particle, and high shear mixing reduces the size of boundary
layer to facilitate transport with the bulk liquid. Automated continuous addition of NCA solid at a
defined and consistent rate allowed the maintenance of the optimal 0.1 equiv. excess over the co-
fed pH10.2 amino acid solution. Nevertheless, with these fast reactions precise control of
stoichiometry may still be problematic. Whilst the L-Ala NCA fine white particulate material was
well behaved (ensuring exclusion of moisture), the L-Val NCA particles were more cohesive,
requiring regular tapping of the hopper, and making its addition less consistent. Clearly, the
material properties of other NCAs may affect the ease with which they can be added. Consideration
of the base stoichiometry shows the relative rates of NCA reaction with amino acid to base
hydrolysis are about 2:1. Based on steady-state hydroxide and amino acid concentrations of
ACS Paragon Plus Environment
16

Page 17 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
0.16mM and 0.1M, and assuming both reactions are first order in NCA, the rate constant for the
reaction with hydroxide is 373 times that of the amino acid. Minimising hydrolysis is difficult,
however if the amino acid and NCA concentrations were increased to 1M at the same hydroxide
concentration, the dipeptide yield might be improved to 94%.43 This would require more rapid
addition of more material and would test the physical boundaries of the system, indicating that
further improvements to the reactor design would be necessary.
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
60
Further work is required to determine the optimal residence time, and as the studies indicate this
may vary for individual NCA and amino acid/peptide reactions. The fast reaction makes it possible
to maintain a high conversion with a shorter residence time, eg. <1 minute, and this would alleviate
further reaction of the product; a smaller volume multi-stage CSTR would improve the efficiency,
however the need to add solids makes this difficult.44 A homogenous reaction would considerably
simplify the equipment and this approach is currently being investigated.
The productivity of the continuous process is much higher than other methods, and this could help
reduce the costs of peptide production,45, 46 however a current problem is isolation of the peptides
from large volumes of salt solutions. Work to develop continuous membrane separation and
purification processes is ongoing. Realistically the flow NCA method is likely to be limited to
tetra- or pentapeptides, but these might coupled by block condensation to form longer chain
peptides.47
The continuous flow synthesis of a series of di- and tri-peptides has been achieved through
understanding the chemical and physical requirements of the reaction and this information used to
design a multi-phasic flow reactor that minimises unwanted side-reactions. The concept of
designing a reactor to meet the physicochemical requirements of a reaction can applied to other
complex systems. A further useful aspect described here is the use of rapid and precise pH control
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 27
1
2
3
4
5
6
to avoid the use of a buffer thereby improving the green metrics of for example biocatalysed
reactions.
7
8
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
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL
Commercially available chemicals used in this work were obtained from Sigma Aldrich and
Fluorochem of reagent or reagent plus grade. L-Alanyl NCA and L-valyl NCA were obtained from
Isochem and were stored under N2 at -20 ᵒC. NMR: Bruker DPX-300 (300 MHz), Bruker Ascend
400 (400 MHz) or Bruker DRX-500 (500 MHz) instruments using D2O solvent. The HPLC
employed an Agilent 1100 Series HPLC with Chemstation software for processing. LCMS was
carried out using a Bruker HCT Ion trap mass spectrometer coupled with Agilent 1100 HPLC.
HRMS was done with a Bruker Maxis Impact spectrometer with Ultimate 3000 UPLC. IR spectra
were determined using a Bruker ALPHA FT-IR spectrometer. Melting points (Mp) with Stuart
Melting Point Aparatus SMP30. Optical rotations were determined with a Schmidt and Haensch
Polartronic H 352 with 1 dm cell. Mass Directed Automated preparative chromatography (MDAP)
was done using Agilent Technologies 2120 Quadrupole mass spectrometer with Agilent
Technologies 1260 Infinity HPLC.
A detailed description of the continuous flow equipment built and used in the study is shown in
photos, diagrams and described in the ESI Section 7.3. The overflow pipe in the reactor was set at
the 300 mL level. To the reactor was added 300 mL of the reaction solvent (0.1M Na2B4O7 aq. or
water/base) and the pH was adjusted to pH10.2 by addition of base as necessary. The solvent was
stirred (4000 rpm) and cooled to 0 °C. Once at 0°C, to the reactor was then fed 0.2 M, pH 10.2
aqueous amino acid solution (41 mL/min, 8.2 mmol/min.), the required NCA (9 mmol/min.) and
base, if required, at a rate necessary to maintain the reaction solution at pH 10.2. The reaction was
ACS Paragon Plus Environment
18

Page 19 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
stirred at 4000 rpm and the temperature maintained between 0-2 °C. The reaction solution was
collected via the overflow pipe. During the reaction, the reaction temperature, pH and flow rate of
base were recorded each minute. After 7 min. (1 residence time), 300 mL (1 reactor volume) of
solution had been collected. The collection vessel was replaced and the process repeated until the
required number of reactor volumes had been collected. Once collected, each reactor volume was
acidified with 3M sulfuric acid to allow decarboxylation of the carbamate intermediate. The
products from each individual reactor volume were analysed by NMR and HPLC. For NMR, 1
mL of the reaction solution was dried by rotary evaporation and a portion of the solid was dissolved
in D2O. For the initial reactions, a known mass of NaOAc was added to the NMR sample as an
external standard however results from this were found to be inconsistent and were not used
further. For HPLC, 20 μL of the reaction solution was added to 10 μL of 0.1M biphenyl in MeOH
(external standard). To this was added 500 μL MeOH and 470 μL H2O. Grace Davidson Vydac
C18(218TP) 250 x 4.6 mm, pore size 300 Å, particle size 5 μm, pH 2-7.5; 210 nm; 25 °C column
oven; 10 μL injection volume; 1 mL/min.; solvent A: Water + 0.1% TFA; solvent B: MeCN +
0.1% TFA; Gradient: 0 min 5% B, 16 min 40% B, 18 min 40% B, 20 min 80% B, 27 min 80% B,
28 min 5% B. The retention times of amino acids and peptides were: L-Phenylalanine: 7.2 min; L-
Alanyl-L-phenylalanine: 8.9 min; DL-Alanyl-DL-phenylalanine: 8.9 and 10.6 min; L-Alanyl-L-
alanyl-L-phenylalanine: 14.0 min; L-Valyl-L-phenylalanine: 14.7 min; L-Alanyl-L-leucine: 10.2
min; L-Leucine: 7.0 min; L-Valyl-L-alanyl-L-leucine: 17.3; L-Valyl-L-aspartyl-L-phenylalanine:
18.0 min; L-aspartame: 16.2 min; Biphenyl (external standard): 25.7 min.
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
60
Purification was carried out by mass directed auto preparative chromatography (MDAP) or ion
exchange and flash chromatography. For the former technique, small quantities (5-10 mL) of the
crude reaction solutions were purified to give samples of the desired reaction products for
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 27
1
2
3
4
5
6
7
8
9
characterisation and to act as HPLC standards (see ESI Section 4.5 Peptide products). The MDAP
method employed an Xbridge Prep C18 5 μm OBDTM 19x100 mm. 5-95% MeOH in water with
0.1% formic acid over 8 min., 20 mL/min flow rate. Purification by ion exchange chromatography
was done using Dowex 50WX4 100-200 mesh ion exchange resin prewashed with water. 20 mL
of the crude reaction solution was basified to pH 5.5 with 1 M KOH aq. and loaded onto the
column. The products were eluted with firstly a pH 5.5 aqueous solution, followed by pH 5.7 and
finally pH 6. The fractions collected were analysed by TLC (Si gel, CHCl3: MeOH: NH4OH 18%
aq. 50:40:10, visualisation by UV and PMA dip.) This method was found to successfully remove
salts and tri-, tetra- peptide side products from buffer free reactions. In the reaction to form L-
Alanyl-L-Phenylalanine, the dipeptide was successfully separated from residual phenylalanine
starting material by this method alone, whilst for the other dipeptides this was not possible and
subsequent silica gel flash chromatography was required. This was carried out by loading the
concentrated dipeptide/amino acid mixture onto the silica gel column. The products were then
eluted with DCM: MeOH: NH4OH aq. 18% 30:60:10. The fractions were analysed by TLC using
the same solvent system, and visualised by UV and PMA dip. For L-Val-L-Phe preparation, the
dipeptide was obtained as an enriched mixture along with less phenylalanine than by ion exchange
chromatography alone.
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
60
Characterisation data for the products is as follows.
L-Alanyl-L-Phenylalanine. Mp 242-243ᵒC; [α]D20 +39.3 (589 nm, c 2, water) (lit.48 [α]D25 + 38.8
(L,L)); IR νmax 3229.8, 2923.7, 1668.0, 1522.6, 1222.5, 1076.8, 739.4, 697.5, 608.7 cm-1; 1H NMR
(500 MHz, D2O) 1H NMR (501 MHz, Deuterium Oxide) δ 7.43 (t, J = 7.3 Hz, 3H, CHAr), 7.38 –
7.31 (m, 4H, CHAr), 4.65 (dd, J = 8.9, 5.6 Hz, 1H, CHCH2Ar), 4.03 (q, J = 7.1 Hz, 2H, CHCH3),
3.29 (dd, J = 14.0, 5.6 Hz, 1H, CHaHbAr), 3.09 (dd, J = 14.1, 9.0 Hz, 2H, CHaHbAr), 1.53 (d, J =
ACS Paragon Plus Environment
20

Page 21 of 27
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
13
7.1 Hz, 3H, CH3); C NMR (125 MHz, D2O) 175.78 (C=O), 170.37 (C=O), 137.11 (CiHAr),
129.30 (2 x CoHAr), 128.83 (2 x CmHAr), 127.19 (CpHAr), 55.36 (CHCH2), 49.01 (CHCH3), 36.78
(CH2), 16.49 (CH3); LRMS m/z (ESI) 237.3 (M+ + H); HRMS m/z (ESI): M+ + H, 237.12342
C12H17N2O3 requires M 237.12337.
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
60
20
L-Alanyl-L-Leucine. Mp 226-227ᵒC; [α]D -30.2 (589 nm, c 1, water); IR νmax 3218.1, 3067.3,
2956.8, 1667.4, 1520.7, 1049.5, 606.6 cm-1; 1H NMR (500 MHz, D2O) δ 4.42 (t, J=7.3 Hz, 1H,
CHCH2CH(CH3)2), 4.15 (q, J=7.1 Hz, 1H, CHCH3), 1.73-1.72 (m, 3H, CH2 and CH(CH3)2
overlapping), 1.60 (d, J=7.0 Hz, 3H, CH3), 0.97 (dd, J=17.8, 6.3 Hz, 6H, CH(CH3)2); 13C NMR
(125 MHz, D2O) 176.82 (C=O), 170.85 (C=O), 52.20 (CHCH2(CH3)2), 49.02 (CHCH3), 39.41
(CH2), 24.57 (CH(CH3)2, 22.22 (CH3), 20.72 (CH3), 16.57 (CH3); LRMS m/z (ESI) 203.4 (M+ +
H); HRMS m/z (ESI): M+ + H, 203.13897 C9H19N2O3 requires M 203.13902.
L-Valyl-L-Phenylalanine. Mp 233-234ᵒC; [α]D20 +70 (589 nm, c 1, water); IR νmax 3218.1, 3067.3,
2956.8, 1667.4, 1520.7, 1049.5, 606.6 cm-1; 1H NMR (500 MHz, D2O) δ7.44-7.41 (m, 2H, CHAr),
7.38 – 7.35 (m, 3H, CHAr), 4.67 (dd, J = 8.6 Hz, 1H, CHCH2Ar), 3.80 (d, J = 5.6 Hz, 1H,
CHCH(CH3)2), 3.27 (dd, J = 14.1, 5.8 Hz, 1H, CHaHbAr), 3.10 (dd, J = 14.1, 8.7 Hz, 1H,
CHaHbAr), 2.28 – 2.18 (m, 1H, CH(CH3)2), 1.05 – 1.01 (m, 6H, (CH3)2); C NMR (125 MHz,
13
D2O) 175.78 (C=O), 168.93 (C=O), 137.07 (CiHAr), 129.29 (2 x CoHAr), 128.85 (2 x CmHAr),
127.20 (CpHAr), 58.50 (CHCH2CH(CH3)2), 55.54 (CHCH(CH3)2), 36.88 (CH2), 30.09
(CH(CH3)2), 17.69 (CH3CHCH3), 16.69 (CH3CHCH3); LRMS m/z (ESI) 265.3 (M+ + H); HRMS
m/z (ESI): M+ + H, 265.15510 C14H21N2O3 requires M 265.15467.
20
L-Alanyl-L-Alanyl-L-Phenylalanine. Mp 209-210 ᵒC; [α]D -13.6 (589 nm, c 1, water); IR νmax
3261.8, 3063.9, 2935.8, 1632.8, 1529.9, 1077.7, 958.5, 696.5, 575.8 cm-1; H NMR (500 MHz,
1
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 27
1
2
3
4
5
6
7
8
9
D2O) δ7.42-7.30 (m, 2H, CHAr), 4.63-4.54 (m, 1H, CHCH2Ar), 4.34 (q, J = 7.2 Hz, 1H, CHCH3),
4.08 (q, J = 7.1 Hz, 1H, CHCH3), 3.24 (dd, J = 13.8, 5.4 Hz, 1H, CHaHbAr), 3.08 (dd, J = 13.8,
7.9 Hz, 1H, CHaHbAr), 1.51 (d, J = 7.0 Hz, 3H, CH3), 1.37 (d, J = 7.1 Hz, 3H, CH3); 13C NMR
(125 MHz, D2O) 176.13 (C=O), 173.87 (C=O), 170.50 (C=O), 137.00 (CiHAr), 129.47 (2 x
CoHAr), 128.74 (2 x CmHAr), 127.10 (CpHAr), 55.13 (CHCH2Ar), 49.82 (CHCH3), 49.00
(CHCH3), 37.17 (CH2), 16.69 (CH3), 16.65 (CH3); LRMS m/z (ESI) 308.4 (M+ + H); HRMS m/z
(ESI): M+ + H, 308.16069 C15H22N3O4 requires M 308.16048.
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
60
L-Valyl-L-Alanyl-L-Leucine. Mp 341-342 ᵒC; [α]D20 -93.4 (589 nm, c 1, water); IR νmax 3295.1,
1
3065.6, 2961.1, 1644.7, 1541.2, 1232.1, 1160.8, 1044.2, 605.1 cm-1; H NMR (500 MHz, D2O)
δ4.48 (q, J = 7.2 Hz, 1H, CHCH3), 4.41-4.35 (m, 1H, CHCH2), 3.85 (d, J = 5.9 Hz, 1H,
CHCH(CH3)2), 2.30-2.21 (m, 1H, CH(CH3)2), 1.74-1.69 (m, 3H, CH2 and CH(CH3)2 overlapping),
1.45 (d, J = 7.2 Hz, 3H, CH3), 1.08-1.06 (m, 6H, (CH3)2, 0.98 (d, J = 6.3 Hz, 3H, (CH3)2, 0.93 (d,
J = 6.3 Hz, 3H, (CH3)2); 13C NMR (125 MHz, D2O) 177.11 (C=O), 174.27 (C=O), 168.89 (C=O),
58.45 (CHCH(CH3)2), 52.04 (CHCH2(CH3)2), 49.58 (CHCH3), 39.69 (CH2), 30.10
(CH2CH(CH3)2), 24.52 (CHCH(CH3)2), 22.32 (CH3), 20.71 (CH3), 17.65 (CH3), 16.94 (CH3),
16.50 (CH3); LRMS m/z (ESI) 302.4 (M+ + H); HRMS m/z (ESI): M+ + H, 302.20749 C14H28N3O4
requires M 302.20743.
20
L-Valyl-L-Aspartyl-L-Phenylalanine. Mp 184-186 ᵒC; [α]D -5.6 (589 nm, c 1, water); IR νmax
3193.3, 3066.0, 1713.8, 1660.3, 1517.5 1030.3, 581.8 cm-1; 1H NMR (500 MHz, D2O) δ7.37-7.14
(m, 5H, CHAr in tripeptide and 2H CHAr in impurity), 4.69-4.67 (m, 1H, CHCH2CO2H), 4.49
(dd, J = 8.0, 5.3 Hz, CHCH2Ar), 4.40-4.38 (m, 0.4H, impurity), 4.14-4.12 (m, 0.4H, impurity),
3.69 (d, J = 5.9 Hz, 1H, CHCH(CH3)2), 3.16 (ddd, J = 38.2, 13.8, 4.6 Hz, 1.4H, CHaHbCO2H and
impurity), 2.98-2.95 (m, 1.4H, CHaHbCO2H and impurity), 2.77 (dd, J = 16.7, 5.0 Hz, 1H,
ACS Paragon Plus Environment
22

Page 23 of 27
Organic Process Research & Development
1
2
3
4
CHaHbAr), 2.64 (dd, J = 16.8, 8.8 Hz, 1H, CHaHbAr), 2.10-2.07 (m, 0.5H, impurity), 2.04-1.99
5
6
13
(m, 1H, CH(CH3)2), 0.85 (t, J = xx Hz, 6H, CH(CH3)2); C NMR (125 MHz, D2O) Complex
7
8
9
spectra obtained due to unidentified impurity. Peaks could not be assigned; LRMS m/z (ESI) 380.4
(M+ + H); HRMS m/z (ESI): M+ + H, 380.18241 C18H26N3O6 requires M 380.18161.
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
60
ASSOCIATED CONTENT
Supporting Information. This material is available free of charge via the Internet at
http://pubs.acs.org.
Information on stoichiometry, heat flow, equipment design, green metrics, analytical and
experimental details, product characterization NMR and HPLC spectra.
AUTHOR INFORMATION
Corresponding Author
* E-mail: j.blacker@leeds.ac.uk
ACKNOWLEDGEMENT
The research leading to these results has received funding from the Innovative Medicines Initiative
Joint Undertaking project CHEM21 under grant agreement number 115360, the European Union’s
Seventh Framework Programme (FP7/2007-203) and EFPIA companies’ in kind contribution.
Notes
The authors declare no competing financial interest.
REFERENCES
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 27
1
2
3
4
5
6
1.
Santos, S.; Torcato, I.; Castanho, M. A. R. B., Biomedical applications of dipeptides and
tripeptides. Peptide Science 2012, 98, 288-293.
2. Falco, A.; Ortega-Villaizan, M.; Chico, V.; Brocal, I.; Perez, L.; Coll, J. M.; Estepa, A.,
7
8
9
Antimicrobial peptides as model molecules for the development of novel antiviral agents in
aquaculture. Mini Rev. Med. Chem. 2009, 9, 1159-64.
3.
Mine, Y. L.-C., E.; Jiang, B, Bioactive Proteins and Peptides as Functional Foods and
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
60
Nutraceuticals. Wiley-Blackwell: USA.
4.
5.
Scanlon, S.; Aggeli, A., Self-assembling peptide nanotubes. Nano Today 2008, 3, 22-30.
Thayer, A. M., Making Peptides At Large Scale. Chem. Eng. News May 30 2011 pp 21 -
25.
6.
Felix, A.; Moroder, L.; Toniolo, C., Synthesis of Peptides and Peptidomimetics. Thieme
Stuttgart: New York, 2002; Vol. E22a, p 1-901.
7. Merrifield, R. B., Solid Phase Synthesis (Nobel Lecture). Angew. Chem. Int. Ed. 1985,
24, 799-810.
8.
Methods in Enzymology: Solid-phase peptide synthesis. Academic Press: San Diego,
California, London, 1997; Vol. 289.
9.
Am. Chem. Soc. 1963, 85, 2149-2154.
10. König, W.; Geiger, R., Eine neue Methode zur Synthese von Peptiden: Aktivierung der
Merrifield, R. B., Solid Phase Peptide Synthesis. I. The Synthesis of a Tetrapeptide. J.
Carboxylgruppe mit Dicyclohexylcarbodiimid unter Zusatz von 1-Hydroxy-benzotriazolen.
Chem. Ber. 1970, 103, 788-798.
11.
12.
Harada, T.; Irino, S.; Kunisawa, Y.; Oyama, K. EP768384, 1997.
Yagasaki, M.; Hashimoto, S., Synthesis and application of dipeptides; current status and
perspectives. Appl. Microbiol. Biotechnol. 2008, 81, 13-22.
13. Tsuda, Y.; Okada, Y., Solution-Phase Peptide Synthesis. In Amino Acids, Peptides and
Proteins in Organic Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA2010; pp 201-251.
14. Heinzel, W.; Verlander, M. S., Peptide Synthesis. In Ullmann's Encyclopedia of
Industrial Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA2000.
15. Baeshen, N. A.; Baeshen, M. N.; Sheikh, A.; Bora, R. S.; Ahmed, M. M. M.; Ramadan,
H. A. I.; Saini, K. S.; Redwan, E. M., Cell factories for insulin production. Microbial Cell
Factories 2014, 13, 141.
16.
Feliu, J. A.; Mas, C. d.; Lopez-Santin, J., Enzyme and Microbial Technology. 1995, 17,
882-887.
17.
Ulijn, R. V.; Baragana, B.; Halling, P. J.; Flitsch, S. J., Journal of the American Chemical
Society 2002, 124, 10988-10989.
18.
Altreuter, D. H.; dordick, J. S.; Clark, D. S., Biotechnology and Bioengineering 2003, 81,
809-817.
19.
622.
20.
21.
Ulijn, R. V.; Bisek, N.; Flitsch, S. L., Organic & Biomolecular Chemistry 2003, 1, 621-
Bray, B. L., Nature Reviews Drug Discovery 2003, 2, 587-593.
Nakanishi, K.; Takeuchi, A.; Matsuno, R., Long-term continuous synthesis of aspartame
precursor in a column reactor with an immobilized thermolysin. Applied microbiology and
biotechnology 1990, 32, 633-636.
22.
Bergmann, M.; Fraenkel-Conrat, H., THE ENZYMATIC SYNTHESIS OF PEPTIDE
BONDS. Journal of Biological Chemistry 1938, 124, 1-6.
ACS Paragon Plus Environment
24

Page 25 of 27
Organic Process Research & Development
1
2
3
4
5
6
23.
Hirschmann, R., Synthesis of peptides in aqueous medium. VII. Preparation and use of
2,5-thiazolidinediones in peptide synthesis. The Journal of Organic Chemistry 1971, 36, 49-59.
24. Veber, D. F.; Hirschmann, R.; Denkewalter, R. G., Synthesis of peptides in aqueous
7
8
9
medium. VI. Synthesis of an unsymmetrical cystine peptide fragment of insulin. The Journal of
Organic Chemistry 1969, 34, 753-755.
25.
enzyme. I. Objective and strategy. Journal of the American Chemical Society 1969, 91, 502-503.
26. Hirschmann, R.; Strachan, R. G.; Schwam, H.; Schoenewaldt, E. F.; Joshua, H.;
Denkewalter, R. G.; Veber, D. F.; Holly, F. W.; Hirschmann, R., Total synthesis of an
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
60
Barkemeyer, B.; Veber, D. F.; Jr, W. J. P.; Jacob, T. A.; Beesley, T. E.; Denkewalter, R. G.,
Controlled synthesis of peptides in aqueous medium. III. Use of Leuch's anhydrides in the
synthesis of di-peptides. Mechanism and control of side reactions. The Journal of Organic
Chemistry 1967, 32, 3415-3425.
27.
Denkewalter, R. G.; Schwam, H.; Strachan, R. G.; Beesley, T. E.; Veber, D. F.;
Schoenewaldt, E. F.; Barkemeyer, H.; Paleveda, W. J.; Jacob, T. A.; Hirschmann, R., The
Controlled Synthesis of Peptides in Aqueous Medium. I. The Use of α-Amino Acid N-
Carboxyanhydrides. Journal of the American Chemical Society 1966, 88, 3163-3164.
28.
Ager, D. J.; Pantaleone, D. P.; Henderson, S. A.; Katritzky, A. R.; Prakash, I.; Walters,
D. E., Commercial, Synthetic Nonnutritive Sweeteners. Angewandte Chemie International
Edition 1998, 37, 1802-1817.
29.
30.
Kircher, K.; Bendt, M.; Zahn, H., Liebigs Annalen der Chemie 1980, 275-284.
J. F. Douglas, J. M. G., J. A. Swaffield, Fluid Mechanics. 3rd ed.; Longman Pub. Group:
White Plains, New York, 1995.
31. Kricheldorf, H. R., α-Amino acid-N-Carboxy-Anhydrides and related heterocycles.
Syntheses, properties, peptide synthesis, polymerisation. Springer-Verlag: Berlin Heidelberg,
Germany, 1987.
32.
Solution. II. Journal of the American Chemical Society 1957, 79, 2159-2160.
33. Bartlett, P. D.; Jones, R. H., A Kinetic Study of the Leuchs Anhydrides in Aqueous
Solution. I. Journal of the American Chemical Society 1957, 79, 2153-2159.
Bartlett, P. D.; Dittmer, D. C., A Kinetic Study of the Leuchs Anhydrides in Aqueous
34.
35.
Caplow, M., Journal of the American Chemical Society 1968, 90, 6795-6803.
Johnson, S. L.; Morrison, D. L., Journal of the American Chemical Society 1972, 94,
1323-1334.
36.
Knudsen, K. R.; Ladlow, M.; Bandpey, Z.; Ley, S., Fully Automated Sequence-Specific
Synthesis of α-Peptides Using Flow Chemistry. Journal of Flow Chemistry 2014, 4, 18-21.
37.
Fuse, S.; Mifune, Y.; Takahashi, T., Efficient Amide Bond Formation through a Rapid
and Strong Activation of Carboxylic Acids in a Microflow Reactor. Angewandte Chemie
International Edition 2014, 53, 851-855.
38.
NCAs are available from a number of suppliers. Those used in this study were purchased
from Isochem S.A.S., France
39.
NCAs are available from a number of suppliers. Those used in this study were purchased
from Isochem
40.
Gait, R. J.; Langlois, J. R.; Williams, R. E., Use of N-Carboxyanhydrides in the Synthesis
of Dipeptides. The Improvement of Isolated Yields by the Facile Removal of Buffer
Components. Canadian Journal of Chemistry 1972, 50, 299-303.
41.
http://www.crchandbook.com/ (03/04/17),
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 27
1
2
3
4
5
6
42.
McElroy, C. R.; Constantinou, A.; Jones, L. C.; Summerton, L.; Clark, J. H., Towards a
holistic approach to metrics for the 21st century pharmaceutical industry. Green Chemistry 2015,
17, 3111-3121.
7
8
9
43.
Based on 62% peptide and 37% hydrolysis with 100mM amino acid and 0.16mM (pH
10.2) hydroxide = 6.2/2312=372. With 1M amino acid the relative rates are 6.2:0.37 giving a
possible 94% dipeptide yield.
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
60
44.
Chapman, M. R.; Kwan, M. H.; King, G.; Jolley, K. E.; Hussain, M.; Hussain, S.;
Salama, I. E.; González Nino, C.; Thompson, L. A.; Bayana, M. E.; Clayton, A.; Nguyen, B. N.;
Turner, N. J.; Kapur, N.; Blacker, A. J., A Simple and Versatile Laboratory Scale CSTR for
Multiphasic Continuous-Flow Chemistry and Long Residence Times. Org. Proc. Res. Dev. 2017.
45.
46.
47.
peptide ligations and the total synthesis of proteins. Current Opinion in Chemical Biology 2014,
22, 115-121.
48.
Thayer, A. M., Chemical and Engineering News May 30th 2011, 89, 21-25.
Kelley, W. S., Biotechnology 1996, 14, 28-31.
Harmand, T. J. R.; Murar, C. E.; Bode, J. W., New chemistries for chemoselective
Lemieux, R. U.; Barton, M. A., Peptide Conformations. I. Nuclear Magnetic Resonance
Study of the Carbamate Reaction of Amino Acids and Peptides. Canadian Journal of Chemistry
1971, 49, 767-776.
ACS Paragon Plus Environment
26

Page 27 of 27
Organic Process Research & Development
1
2
3
4
5
6
Organic Process Research & Development
Msc: op-2017-002143
7
8
9
The following graphic will be used for the TOC:
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
60
ACS Paragon Plus Environment