Peptidyl-Prolyl Model Study: How Does the Electronic Effect Influence the Amide Bond Conformation?

Article abstract of DOI:10.1021/acs.joc.7b00803

The triple-helical structure of collagen, the most abundant protein in animal bodies, owes its stability to post-translationally installed hydroxyl groups at position 4 of prolyl residues. To shed light on the nature of this phenomenon, we have examined the influence of the 4-substituent on the amide isomerism in peptidyl-prolyl analogues. The rigid bicyclic skeleton of 2,4-methanoprolines allowed us to follow the through-bond impact of the substituent group (electronic effect) without the side-chain conformation being affected by a stereoelectronic effect. These proline analogues were prepared by [2 + 2] photocycloaddition of (2-allylamino)acrylic acid derivatives. Subsequent pK_a studies demonstrated a remarkable electronic effect of the 4-fluorine substitution, while the effect of the 4-methyl group was negligible. The trans/cis amide ratio was measured in model compounds under low temperature conditions. The observed prevalence for a trans-amide is extraordinary, and in this regard, 2,4-methanoproline is closer to primary α-amino acids than to proline. At the same time the amide rotation velocities were 3-4 orders of magnitude higher when compared to N-acetylprolyl. Finally, our results indicate that the electronic effect of the 4-substituent only affects the kinetics of the amide isomerization but not the thermodynamic prevalence for the trans-rotamer.

Full text of DOI:10.1021/acs.joc.7b00803

Article
A peptidyl-prolyl model study: How does the electronic
effect influence the amide bond conformation?
Pavel K. Mykhailiuk, Vladimir Kubyshkin, Thorsten Bach, and Nediljko Budisa
J. Org. Chem., Just Accepted Manuscript • Publication Date (Web): 07 Jun 2017
Downloaded from http://pubs.acs.org on June 9, 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.
The Journal of Organic Chemistry 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 42
The Journal of Organic Chemistry
1
2
3
4
5
6
A Peptidyl-Prolyl Model Study: How Does the Electronic Effect Influence the
7
8
9
Amide Bond Conformation?
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
Pavel K. Mykhailiuk,*†, ‡ # Vladimir Kubyshkin,* § # Thorsten Bach,‖ Nediljko Budisa§
† Enamine Ltd., Chervonotkatska 78, 01103 Kyiv, Ukraine, www.enamine.net
‡ Taras Shevchenko National University of Kyiv, Chemistry Department, Volodymyrska 64, 01601 Kyiv,
Ukraine
§ Institute of Chemistry, Technical University of Berlin, 10623 Berlin, Germany
‖ Lehrstuhl für Organische Chemie I, Technische Universität München, Lichtenbergstr. 4, 85747 Garching,
Germany
# P.K.M. and V.K. contributed equally to this work.
E-mails: Pavel.Mykhailiuk@gmail.com, Pavel.Mykhailiuk@mail.enamine.net (PKM),
Kubyshkin@win.tu-berlin.de (VK)
The authors declare no competing financial interest.
1
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 42
1
2
3
4
TOC graphic
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
2
ACS Paragon Plus Environment

Page 3 of 42
The Journal of Organic Chemistry
1
2
3
4
Abstract
5
6
7
8
The triple-helical structure of collagen, the most abundant protein in animal body, owes its stability to
post-translationally installed hydroxyl groups at position 4 of prolyl residues. To shed light on the nature
of this phenomenon we have examined the influence of the 4-substituent on the amide isomerism in
peptidyl-prolyl analogues. The rigid bicyclic skeleton of 2,4-methanoprolines allowed us to follow the
through-bond impact of the substituent group (electronic effect), without the side-chain conformation
being affected by a stereoelectronic effect. These proline analogues were prepared by [2+2]
photocycloaddition of (2-allylamino)acrylic acid derivatives. Subsequent pKa studies demonstrated a
remarkable electronic effect of the 4-fluorine substitution, while the effect of the 4-methyl group was
negligible. The trans/cis amide ratio was measured in model compounds under low temperature
conditions. The observed prevalence for a trans-amide is extraordinary, and in this regard 2,4-
methanoproline is closer to primary -amino acids than to proline. The amide rotation barrier was also
found to be significantly reduced: as equivalent to 3-4 orders of magnitude higher isomerization velocities
when compared to N-acetyl-prolyl. Finally, our results indicate that the electronic effect only affects the
kinetics of the amide isomerization, but not the thermodynamic prevalence for the trans-rotamer.
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
3
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 42
1
2
3
4
Introduction
5
6
7
8
Collagen is the most abundant protein in mammals, and its malfunctions are related to a
variety of diseases.1 Perhaps, the historically best known among these diseases is scurvy, which
is caused by a lack of vitamin C, essential for maintaining a correct hydroxylation rate of
repetitive prolyl-prolyl-glycyl motifs in procollagen.2,3 This post-translational modification
yields (4R)-hydroxyproline (1, Chart 1) residues in positions preceding glycine. As a result, the
stability of triple helical arrangements increases and the melting point of collagen is adjusted
above body temperature, ensuring the correct integrity of connective tissues.4
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
Chart 1 Structures of (4R)-hydroxyproline (1), proline (2) and diastereomeric 4-fluoro- (2a, 2b) and 4-
methylprolines (2c, 2d).
The essential role of prolyl hydroxylation had not been well understood until the late 1990s,
when Raines and colleagues discovered that incorporation of (4R)-fluoroproline (2a) into
repetitive collagen mimicking sequences results in even higher thermal stability as compared to
analogous sequences containing hydroxyproline (1).5 It therefore became apparent that neither
hydrogen bond nor water network stabilization is promoted by prolyl hydroxylation in collagen,
but that rather an electron withdrawing substituent enhances the stability of the structure due to
4
ACS Paragon Plus Environment

Page 5 of 42
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
its electronic effect. The latter can be conceived as a through-bond interaction of a side chain
substituent to the backbone groups, and it is manifested by a significant reduction of the amide
bond isomerization barriers (Scheme 1A).6
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
This simple hypothesis was later confronted with the fact that incorporation of (4S)-
fluoroproline (2b) in the glycine-preceding positions decreases the stability of the triple helical
motif dramatically.7 Another remarkable finding was that the incorporation of 4-methylprolines
(2c, 2d) into collagen-mimicking sequences impacts the triple helix stability to a similar extent
as do 4-hydroxy- or 4-fluoroprolines.8 Yet, the effect of methylprolines exhibits a reverse chiral
bias as compared to fluoroprolines: it is the (4S)-isomer (2d) which stabilizes the triple helix
when incorporated in front of glycine residues in the sequence, whereas (4R)-isomer (2c)
exhibits a destabilizing effect. Based on these observations, the stabilizing effect of the (4R)-
fluoroproline was assigned to a stereoelectronic effect,9 which is based on the fluorine gauche-
effect.10 A fluorine gauche conformation stabilizes the C4-exo pucker in 2a (Scheme 1B), which
promotes an n→* interaction between an amide oxygen lone pair and the upstream carbonyl
group. The latter interaction is eventually responsible for a thermodynamic stabilization of the
trans amide bond. Stabilization of both the prolyl exo-pucker and the trans-amide bond are
critical for the folding stability and packing of collagen triple helical chains.11
5
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 42
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. Schematic illustration of the electronic (A) and stereoelectronic effect (B) on the (4R)-fluoroproline
structure.a
a The numeric data are taken from ref.12 (proline and analogues) and ref.10 (gauche-effect).
The discovery of the stereoelectronic effect provided an essential understanding of the
polyproline fold, and this concept was later employed for the construction of clickable,13 pH-
switchable,14 and hydrophobic15 scaffolds based on stable all-trans amide polyproline II type
6
ACS Paragon Plus Environment

Page 7 of 42
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
helical structures.16,17 The ubiquity of the n→* interaction in protein structures has been
demonstrated,18 and it was shown that the n→* interaction is competitive with hydrogen
bonding, since both require an electron density donation from the carbonyl oxygen atom.19
Numerous proline-to-fluoroproline substitutions have been reported, and these showed adverse
effects on the folding and stability of proteins and protein domains due to their impact on the
trans/cis amide isomerism.20,21 Furthermore, effects onto the thermodynamic stability of amide
bond conformers have been identified for a variety of 4-substituted prolines,22 providing an
important toolbox for the functionalization of proteins and peptide motifs.
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
Nevertheless, it is still not possible to completely discriminate between electronic and
stereoelectronic effects based on the existing data on proline analogues. For instance, we recently
demonstrated that the carboxylic group in N-acetylated (4R)-fluoroproline (Ac-2a) is
considerably more acidic than in the respective (4S)-isomer (Ac-2b).23 The higher
electrophilicity of the carboxylic group in 2a can also be inferred from the remarkably high
velocity of the peptidyl transfer reaction with this amino acid measured on the ribosome (about
150-fold increase compared to proline).24 Since n→π* interaction is a donation, one could expect
that a more electrophilic C-terminal carboxyl group in 2a may enhance the strength of this
interaction, and this can lead to some trans-amide stabilization following the electronic effect in
addition to the existent stereoelectronic effect.
7
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 42
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
Chart 2 Model compounds that enable for elucidation of the electronic effect on prolyl amide isomerisation.
It is therefore desirable to discriminate between the two possible stabilization effects. An
interesting approach, which addressed this issue was previously presented by Krow, Raines, and
co-workers.25 In particular, they prepared compound Ac-3b-OMe and demonstrated that its
amide equilibrium constant was nearly identical to that of parent non-fluorinated compound Ac-
3-OMe (Chart 2). Although, this finding provided an important indication for the insignificance
of an electronic effect on the amide rotameric ratio, it should be kept in mind that
conformationally restricted compound 3b has a rigidly fixed antiperiplanar orientation of the
fluorine substituent with respect to the nitrogen atom. Compound Ac-3a-OMe which would
mimic the gauche-conformation as seen in 4-fluoroprolines was not studied. In addition, no
numerical values (such as pKa) were obtained and it remains therefore unclear, to which extent
the 4-substituent influences electronic properties of the amino acid functional groups in this
model system.
This motivated us to design a model study based on the 2,4-methanoproline structure (Chart
2). We assumed that the rigid 2-azabicyclo[2.1.1]hexane skeleton would allow for observation of
8
ACS Paragon Plus Environment

Page 9 of 42
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
the pure electronic effect of a 4-substituent, since the amino acids are devoid of any puckering
transition and any chiral bias. We therefore performed the synthesis of the model compounds
Ac-4/5/6-OMe and report herein on the properties of the amide bond formed in these structures.
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
9
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 42
1
2
3
4
5
6
Results and Discussion
Synthesis
7
8
9
While parent 2,4-methanoproline (4) is well known, and reported to be an anti-insect
metabolite produced in some plants,26,27 methanoproline analogues 5 and 6 had to be synthesized.
Recent advances in photochemical transformations allowed for the preparation of these
compounds,28,29 and we have optimized a gram-scale synthesis of derivatives bearing electron
withdrawing (fluorine atom) and donating (methyl group) substituents in position 4 of the
bicyclic skeleton.
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
In 2009, we already reported on the synthesis of amino acid 5 on a small scale (20 mg).30
The most problematic step of the synthesis was the allylation of 7 to 9, and only a 25% yield was
achieved. We therefore performed an optimization of this step, and found that the use of purified
reagent 831 and an increase of its amount to 3.0 equiv. led to a much better yield of the target
amide 9 (79%, Scheme 2). Lower amounts of alkylating agent gave significantly lower yields,
whereas a larger excess of 8 did not result in any further improvement. Subsequently, the
intramolecular crossed [2+2] photocycloaddition of alkene 9 to proline derivative Bz-5-OMe was
smoothly performed under 366 nm irradiation using benzophenone as the sensitizer.
Saponification of the ester group in Bz-5-OMe produced free acid Bz-5, which was transformed
into the target amino acid 5·HCl using acidic hydrolysis on gram scale.
10
ACS Paragon Plus Environment

Page 11 of 42
The Journal of Organic Chemistry
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 2. Optimized synthesis of the amino acid 5∙HCl.
Synthesis of amino acid 6 was performed analogously (Scheme 3). Alkylation of
compound 7 with 3.0 equiv. of methallyl bromide (10) afforded the product 11 in 76% yield.
Alternatively, diene 11 was synthesized from methyl pyruvate (12) and amine 13 in 60% yield.
The key transformation – intramolecular [2+2] photocycloaddition of 11 – was easily performed
at 366 nm giving the bicyclic compound Bz-6-OMe in 94% yield. Simultaneous hydrolysis of
both amide and ester groups in Bz-6-OMe gave the target amino acid 6∙HCl also on gram scale.
Scheme 3. Synthesis of the amino acid 6∙HCl.
With the amino acids in hand, the model compounds for further physicochemical
characterizations were prepared using standard methods (Scheme 4). In addition, analogous
11
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 42
1
2
3
4
5
6
7
8
9
derivatives of 2-methylproline (2e) were prepared for comparison. 2-methylproline is a well
studied -substituted proline analogue,32,33,34 and it should represent a case of a quaternary
proline for comparison with 2,4-methanoprolines which are also quaternary amino acids.
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 4. Synthesis of the model compounds for physicochemical characterizations.
The electronic effect
The characterization of the electronic effects in 2,4-methanoprolines was then attempted
using numerical parameters. Perhaps the most straightforward measurable value that accesses the
electronic influence of amino acid substituents on the functional groups is the acidity. We
therefore measured acidity constants for the ammonium and for the carboxylic group of 4-
substituted 2,4-methanoprolines. The values we reported previously for 4-substituted prolines are
included for comparison.23,24
12
ACS Paragon Plus Environment

Page 13 of 42
The Journal of Organic Chemistry
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 Acidity of the functional groups as measured by NMR in aqueous medium at 298 K.
Analysis of the obtained values (Figure 1) clearly indicates the absence of any detectable
electronic influence of the 4-methyl group on both functional groups of the examined amino acid.
In contrast to that, a 4-fluorine substituent exhibits a significant electron-withdrawing effect. In
the case of 2,4-methanoprolines, the reduction of the basicity (N-terminal) was by 2.1 pKa units
(Figure 1A), and the increase in the carboxylic acid acidity (C-terminal) was by about 0.6 pKa
units (Figure 1B). Thus, a significant increase of the C-terminal carboxyl electrophilicity can be
expected, and this increase should correlate with an electronic effect on the trans/cis equilibrium.
The amide isomerism
It is commonly accepted that the trans/cis amide propensities for peptidyl-prolyl
modifications can be characterized using simple model compounds. The most extensively
exploited models are methyl esters of N-acetyl amino acids (Ac-AA-OMe, where AA is the
amino acid).11,12 We thus set up to measure trans/cis amide equilibrium values for Ac-4/5/6-OMe
13
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 42
1
2
3
4
in deuterium oxide solution by NMR. In particular, our aim was to examine whether or not the
5
6
7
trans-amide stabilizing n→* interaction was substantially affected by an electronic effect.
8
9
Curiously, the equilibrium population measurements at the standard temperature (298 K)
were hampered by the observation that the barriers of the amide rotation seemed to be extremely
low for 2,4-methanoprolines. Previously, a model compound analysis was performed in the
1980s, where it was reported that 2,4-methanoproline derivatives exist as trans-amides.35
However, no numerical characterization was provided, and it was therefore not clear whether this
observation was a result of a large trans-amide dominance or whether the minor cis-amide could
not be detected due to a fast amide equilibration on the NMR time scale. Our study revealed that
both scenarios appeared to be valid. We found that lowering the temperature of the examined
model substance solutions sufficiently separated the cis-amide resonances from the trans signals
enabling the direct observation of both rotameric species. Subsequently, we determined the
rotameric ratios and the interconversion barriers for the 2,4-methanoproline derivatives (Figure
2).
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
Analysis of the obtained values indicates that for all three examined 2,4-methanoprolines
the prevalence of the trans-amide rotameric form is massive, giving rise to trans/cis ratios in the
range of 140-180 (in aqueous medium at 276 K, Figure 2A). A high trans/cis ratio for 2,4-
methanoproline derivatives was also observed in methanol-d4 solutions (See Table S3 in SI). To
the best of our knowledge, such an extraordinary dominance of one rotameric form has never
been reported for a proline analogue. When converted to free energy, the values found in Ac-
4/5/6-OMe (−11.2 to −11.9 kJ mol−1) are far greater than that for proline (−4.0 ± 0.1 kJ mol−1),
and they already approach the trans-amide preferences of standard -amino acid amides (e.g.
−12.6 ± 0.3 kJ mol−1 for Ac-Gly-OMe at 298 K).23
14
ACS Paragon Plus Environment

Page 15 of 42
The Journal of Organic Chemistry
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 2 Equilibrium and kinetic parameters of the trans/cis amide isomerism in Ac-AA-OMe solution in deuterium
oxide.a
a
Ktrans/cis values were determined at 276 K and converted to the free energy values. The rotational barriers were
determined at 276 (for Ac-4/5/6-OMe) and 310 K (for Ac-2/2e-OMe) and were converted to rate constants using
Eyring equation assuming purely enthalpic energy of isomerization. For details see Tables S3 and S4.
Further analysis of the obtained values indicated almost no influence of the 4-substituent
on the trans/cis ratio. For instance, we expected a significant electronic effect from 4-fluorine
substitution based on the analysis of the pKa values earlier. Indeed, for the Ac-5-OMe we found
a higher prevalence of the trans-rotamer which was however marginal and close to the
measurement error: −0.3 ± 0.3 kJ mol−1 and −0.4 ± 0.3 kJ mol−1 in deuterium oxide and
methanol-d4 solutions, respectively. In contrast to this observation, the stereoelectronic effect in
analogous 4-substituted prolines promotes more significant changes, which extend from −1.2 to
+1.7 kJ mol−1 in relative stabilization of the trans-rotamer (Figure 3). Therefore, the
thermodynamic influence of the pure electronic effect in our model system is minimal at best.
15
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 42
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 3 Manifestation of the stereoelectronic effect in 4-substituted proline analogues (deuterium oxide solution,
298 K). The data taken from ref.12,15 (see Exp. section for details).
Nevertheless, the electronic effect has a detectable influence on the kinetics of the amide
bond isomerization (Figure 2B). Indeed, in the case of 4-fluorine substitution (Ac-5-OMe) the
barriers of the amide rotation are reduced by approx. 3 kJ mol−1 (rate constant enhancement by a
factor ~ 4-5), when compared to both unsubstituted and 4-methylated analogues (Ac-4/6-OMe).
This observation correlates well with the previously observed tendency for 4-substituted prolines:
reduction of the ammonium pKa in the amino acid by one unit resulted in approx. 1 kJ mol−1
reduction of the amide rotational barriers in Ac-AA-OMe models.36
Remarkably, the rotational barriers in all 2,4-methanoproline derivative Ac-4/5/6-OMe
exhibit a vast increase of the amide rotation velocities, and due to this fact the direct observation
of the cis-rotamer at the standard temperature was not possible. When compared to proline (Ac-
2-OMe), the rate increase was about four and three orders of magnitude for cis→trans and
trans→cis transitions, respectively.
16
ACS Paragon Plus Environment

Page 17 of 42
The Journal of Organic Chemistry
1
2
3
4
Structure analysis
5
6
7
8
The unusually high trans-amide preferences and low amide rotation rates observed in
2,4-methanoprolines motivated us to investigate additional structural properties of the model
compounds. To this end, we analyzed X-ray crystal structures of the model substances (Figure
4).37
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 4 X-ray crystal structures for amino acid derivatives.a
a Black – carbon, blue – nitrogen, red – oxygen, green – fluorine. The -angles, and the C=O↔C=O distance of the
putative n→* interaction are highlighted in light blue. Elipsoids contours represent 50% displacement probability.
Firstly, a high trans-amide content should be expected for a quaternary proline analogue
due to apparent steric reasons. Following this argument, an H-to-CH3 substitution in position 2 of
the proline ring (as in 2-methylproline) should potentially result in a higher trans-amide
17
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 42
1
2
3
4
5
6
7
8
9
dominance as compared to H-to-CH2 substitution (as in 2,4-methanoprolines). This conclusion is
however contradicted by our experimental observations, which revealed exactly the opposite
tendency: the trans-amide in Ac-2e-OMe is at least 1 kJ mol−1 less favored than in Ac-4/5/6-
OMe (Figure 2A). This suggests higher steric requirements for amide rotation around 2,4-
methanoproline residues, which may occur due to the rigid conformation of the molecule.
Similarly, extraordinary high trans-amide dominance has been reported in another proline
analogues based on 7-azabicyclo[2.2.1]heptane skeleton, where an -substituent was located at
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 bridgehead carbon atom.38
Indeed, unlike in proline or 2-methylproline, the C-terminal carboxymethyl group in 2,4-
methanoprolines is located at the bridgehead atom, and therefore, it should potentially yield a
zero -angle [C(=O)-N-C-C(=O) torsion]. This implies that the carboxymethyl group should
reside in exactly the same plane as the acetyl substituent, and this placement inevitably features a
steric clash between the two groups, destabilizing the amide ground states in the 2,4-
methanoproline derivatives. This putative steric clash pushes the carboxymethyl group out of
plane, yielding non-zero experimental -angles (±2-41 in crystal, ±11-42 in DFT modelled
structures, see Table S5). The significant steric interference of the C-terminal carboxylate and
the acetyl moiety may explain the observed vast reduction of the amide rotational barriers for
2,4-methanoproline derivatives (by about 15-22 kJ mol−1 when compared to proline).
Further analysis of the structures provides indications for a strong n→* interaction
between the carbonyl groups. For all three 2,4-methanoprolines the found N-C=O↔C=O
distance was about 2.7 Å, which is remarkably lower than the value found in Ac-Pro-OMe (~ 3.2
Å)36 or in the hexaproline crystal structures (2.9 – 3.2 Å),39 but similar to that found in the crystal
18
ACS Paragon Plus Environment

Page 19 of 42
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
structure of Ac-2e-OMe (2.7 Å) or a hexameric structure of an exo-pucker stabilizing proline
analogue (2.7 – 2.9 Å).15 This allows us to conclude that the n→* donation is a prominent
feature of 2,4-methanoprolines and that it is even stronger than in parent prolines. This
conclusion is further supported by the fact, that the acylO…C=Ocarboxyl angle observed in the
crystal structures (99-107°) is very close to the Bürgi-Dunitz angle (107°), at which the
interaction is optimal. The presence of the n→* interaction dictated by the rigid 2,4-
methanoproline scaffold can explain the large preference in favor of the trans-conformation
observed in solution.
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
Finally, the absence of an electronic effect influence can be explained if one considers
possible opposing forces, namely the n→* donative attractive interaction and the n)( repulsion
of the carbonyl groups (Figure 5A).40 If both interactions experience the electron withdrawing
effect of the 4-fluorine substituent to the same extent, the overall effect may be cancelled out.
Indeed, in the salt forms of Ac-4/5/6-O-, in which the electrophilicity of the C-terminal group is
the smallest and the n→* donation should be attenuated, we observed an increase of the trans-
amide stability for 4-fluoro-2,4-methanoproline by about −0.9 ± 0.1 kJ mol−1 (Figure 5B). This
outcome may appear due to the repulsion of the carbonyl groups, which is suppressed by the
electron withdrawing effect of the 4-fluorine substitution.
19
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 42
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 5 n →* donation and n)( repulsion are the two contradicting forces affecting the trans/cis amide
equilibrium. In the salt forms of N-acetyl amino acids, the n →* donation is diminished, however, the 4-fluorine
electronic effect on repulsion force increases the trans-amide content.
20
ACS Paragon Plus Environment

Page 21 of 42
The Journal of Organic Chemistry
1
2
3
4
Conclusions
5
6
7
8
In summary, we found that the 2-azabicyclo[2.1.1]hexane skeleton in 2,4-methanoproline
residues creates a unique steric interference between the C- and N-side carbonyl groups placed in
the same plane. We also conclude that this arrangement enables an effective n→* donative
stabilization in the trans-amide conformation, as seen from the structure analysis. An outcome of
these steric and electronic phenomena is remarkably high trans/cis amide propensities in 2,4-
methanoproline derivatives. The latter were characterized with the values expected for secondary
rather than tertiary amides, which is a significant result for a seemingly trivial one-carbon atom
addition in an amino acid structure.
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
A replacement of a hydrogen atom in position C-4 by a substituent does not impact the
side chain conformation, and subsequently a pure electronic effect was followed by pKa
measurements. While the 4-methyl group does not exhibit such an effect, a 4-fluorine
demonstrates a significant electron-withdrawing effect on both functional groups of the amino
acid. The 4-fluorine substitution exerts a further effect onto amide rotation barriers, giving rise to
a rotation rate enhancement by a factor of ~ 4-5. However, only a marginal influence onto the
trans/cis amide preferences has been found. As a result, we conclude that the electronic effect of
the 4-substituent influences only the kinetic behavior but not the thermodynamic stability of the
trans-amide bond in the studied amino acid model system.
21
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 42
1
2
3
4
5
6
Experimental section
General notes
7
8
9
Solvents were purified according to standard procedures. Other starting materials were
taken from commercial sources at the highest commercial quality and used without further
purification. Column chromatography was performed using Kieselgel Merck 60 (230-400 mesh)
as the stationary phase. Yields refer to chromatographically and spectroscopically (1H NMR)
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
homogeneous material, unless otherwise stated. H, 19F, 13C NMR spectra were recorded at 700,
500 or 400 MHz (1H), 659, 471 or 376 MHz (19F) and 126 or 101 MHz (13C) frequencies.
Chemical shifts are reported in ppm downfield from TMS (1H, 13C) or CFCl3 (19F) according to
the deuterium lock referencing. The N-acetyl amino acid samples were measured in deuterium
oxide with 150 mM KDSO4, the samples were prepared as described earlier.23 13C{1H} dept45,
13
1
13
13C1H HSQC and C1H HMBC experiments were carried out for the assignment of H and C
resonances. Mass spectra were recorded on a LCMS instrument with chemical ionization (CI) or
a GCMS instrument with electron impact ionization (EI). High-resolution mass analysis (HRMS)
was done using electrospray-orbitrap ionization-detection scheme. For the compounds 8, 9, Bz-
5-OMe and 5∙HCl reference spectra were reported in 30.
Synthesis of 4-fluoro-2,4-methanoproline
2-Fluoroprop-2-enyl methanesulfonate (8)
40% solution of 2-fluoroprop-2-en-1-ol in Et2O (5.0 g, 0.026 mol; obtained as
described)30, was diluted with CH2Cl2 (20 ml), and diisopropylethylamine (DIPEA; 6.33 ml,
0.036 mol) was added. The resulting solution was cooled to −30 ºC, and MsCl (2.42 ml, 0.031
mol) was added dropwise. The reaction mixture was allowed to warm to room temperature, and
22
ACS Paragon Plus Environment

Page 23 of 42
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
stirred for additional 1 h. The mixture was washed with water, 10% solution of citric acid, brine,
dried over Na2SO4, and evaporated under reduced pressure. The crude product was purified by
flash column chromatography using pentane/EtOAc = 7/1 as an eluent (Rf = 0.3) to afford the
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
pure compound 8 as a colorless oil. H NMR (500 MHz; CDCl3; Me4Si), δ: 3.09 (s, 3H, CH3),
4.73 (d, 3JH-F = 16.0 Hz, 2H, CH2OMs), 4.61 (dd, J = 46.5, 3.5 Hz, 1H, CHH), 4.97 (dd, J = 15.0,
2.0 Hz, 1H, CHH). 19F NMR (470 MHz; CDCl3; CFCl3), δ: −109.0 (m).
Methyl 2-[benzoyl(2-fluoroprop-2-enyl)amino]acrylate (9)
Amino acid 7 was synthesized following the modified procedure reported previously.30 A
solution of KOtBu (3.30 g, 29.3 mmol, 2.1 eq.) in THF (100 ml) was treated under argon at −78
ºC with a solution of methyl N-benzoyl-3-chloroalaninate 7 (3.30 g, 13.7 mmol, 1.0 eq) in THF
(50 ml) followed by addition of 8 (6.30 g, 40.8 mmol, 3.0 eq). The reaction mixture was allowed
to warm to room temperature, stirred for 12 h, and partitioned between CH2Cl2/H2O (500/100
mL). The organic layer was separated and washed with 10% solution of citric acid, brine, dried
over Na2SO4, and evaporated under reduced pressure. The crude product was purified by flash
column chromatography (pentane/EtOAc = 7/1, Rf = 0.3) to afford the title compound 9 as a
colorless oil (2.81 g, 10.8 mmol, 79% yield). The compound was of 80% purity, and was used in
the next step without additional purification. 1H NMR (500 MHz; CDCl3; Me4Si), δ: 3.67 (s, 3H,
CH3), 4.73 (d, 3JH-F = 13.0 Hz, 2H, NCH2), 4.63 (d, 3JH-F = 49.0 Hz, 1H, CF=CHH), 4.81 (d, 3JH-F
= 17.0, 1H, CF=CHH), 5.68 (s, 1H, NC=CHH), 6.18 (s, 1H, NC=CHH), 7.34-7.45 (m, 3H, Ph),
7.50 (d, 3JH-H = 7.0 Hz, 2H, o-Ph). 19F NMR (470 MHz; CDCl3; CFCl3), δ: −106.3 (m). m/z (CI):
264 (M+1).
Methyl 2-benzoyl-4-fluoro-2-azabicyclo[2.1.1]hexane-1-carboxylate (Bz-5-OMe)
23
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 42
1
2
3
4
Compound 9 (2.65 g, 10.0 mmol) and benzophenone (500 mg) were dissolved in
5
6
degassed CH3CN (500 ml) in a 1L flask. The solution was irradiated at 366 nm during 12 h (the
7
8
9
1
reaction progress was monitored by H NMR). The reaction mixture was concentrated, and the
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
residue was purified by flash column chromatography (hexane/EtOAc = 3/1) to give pure Bz-5-
OMe as a colorless oil (2.40 g, 9.1 mmol, 91% yield). Rf = 0.6 (hexane/EtOAc = 3/1). 1H NMR
(500 MHz; CDCl3; Me4Si), δ: 2.34 (br s, 2H, 2 × CHH), 2.44 (br s, 2H, 2 × CHH), 3.61 (s, 2H,
NCH2), 3.81 (s, 3H, CH3), 7.43 (t, 3JH-H = 7.5 Hz, 2H, Ph), 7.51 (t, 3JH-H = 7.5 Hz, 1H, Ph), 7.73
3
13
2
(d, JH-H = 7.0 Hz, 2H, Ph). C NMR (126 MHz; CDCl3; Me4Si), δ: 47.0 (d, JC-F = 10 Hz, 2 ×
CH2), 52.6 (s, OCH3), 53.9 (d, 2JC-F = 28 Hz, NCH2), 60.4 (d, 3JC-F = 23 Hz, CCO2CH3), 87.8 (d,
1JC-F = 265 Hz, CF), 127.1 (s, CH, Ph), 128.6 (s, CH, Ph), 132.1 (s, CH, Ph), 133.2 (s, C, Ph),
4
19
167.7 (d, JC-F = 11 Hz, CO2CH3), 173.5 (s, NCOPh). F NMR (470 MHz; CDCl3; CFCl3), δ:
−173.1 (s). m/z (CI): 264 (M+1). Anal. calcd for C14H14FNO3: C, 63.87; H, 5.36; N, 5.32. Found:
C, 63.52; H, 5.14; N, 5.67.
2-Benzoyl-4-fluoro-2-azabicyclo[2.1.1]hexane-1-carboxylic acid (Bz-5)
Compound Bz-5-OMe (2.20 g, 8.36 mmol) was dissolved in MeOH (15 ml), and this
solution was added to 2 N NaOH (41.8 mmol, 1.60 g, 5 equiv) upon stirring. Resulting emulsion
was stirred at room temperature for 10 h. The solvent was evaporated, and water (100 ml) was
added. The aqueous layer was washed with CH2Cl2 (3x20 mL) and the organic layer was
discarded. The water phase was acidified with 2 N HCl, and extracted with EtOAc (3x100 mL).
The organic layers were combined, dried over Na2SO4 and concentrated under reduced pressure
to afford the title compound Bz-5 as a yellowish solid (1.93 g, 7.78 mmol, 93% yield). M. p. =
1
176-177 ºC. H NMR (400 MHz; CDCl3; Me4Si), δ: 2.36 (s, 2H, 2 × CHH), 2.54 (br s, 2H, 2 ×
CHH), 3.61 (s, 2H, NCH2), 7.43 (t, 3JH-H = 7.4 Hz, 2H, Ph), 7.52 (t, 3JH-H = 7.4 Hz, 1H, Ph), 7.70
24
ACS Paragon Plus Environment

Page 25 of 42
The Journal of Organic Chemistry
1
2
3
13
2
3
4
(d, JH-H = 7.0 Hz, 2H, Ph). C NMR (101 MHz; CDCl3; Me4Si), δ: 47.7 (d, JC-F = 21 Hz, 2 ×
5
6
7
8
2
3
1
CH2), 54.6 (d, JC-F = 29 Hz, NCH2), 61.9 (d, JC-F = 23 Hz, CCO2H), 87.6 (d, JC-F = 265 Hz,
CF), 128.5 (s, CH, Ph), 128.8 (s, CH, Ph), 132.3 (s, CH, Ph), 133.1 (s, C, Ph), 170.5 (d, 4JC-F = 11
Hz, CO2H), 174.0 (s, NCOPh). 19F NMR (376 MHz; CDCl3; CFCl3), δ: −170.4 (s). m/z (CI): 250
(M+1). Anal. calcd for C13H12FNO3: C, 62.65; H, 4.85; N, 5.62. Found: C, 62.39; H, 4.55; N,
5.83.
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
1-Carboxy-4-fluoro-2-azabicyclo[2.1.1]hexan-2-ium chloride (5∙HCl)
Bz-5 (0.6 g, 2.4 mmol) was dissolved in 6 N HCl (30 ml) upon stirring. The mixture was
refluxed for 4 h. The solvent was removed under reduced pressure, and the crude material was
dissolved in water and hot filtered with activated charcoal. The filtrate was freeze-dried to afford
5∙HCl as a grey solid (0.35 g, 1.9 mmol, 80 % yield). M. p. > 220 ºC. 1H NMR (700 MHz; D2O;
Me4Si), δ: 2.34 (dt, J = 5.6 and 1.5 Hz, 2H, 2 × CHH), 2.63 (br m, 2H, 2 × CHH), 3.46 (s, 2H,
NCH2). 13C NMR (176 MHz; D2O; Me4Si), δ: 44.7 (d, 2JC-F = 21 Hz, 2 × CH2), 46.2 (d, 2JC-F = 30
Hz, NCH2), 60.7 (d, 3JC-F = 22 Hz, NCCOOH), 87.6 (d, 1JC-F = 265 Hz, CF), 168.3 (d, 4JC-F = 10
Hz, COOH). 19F NMR (659 MHz; D2O; CFCl3), δ: −170.5 (s). m/z (CI): 146 (M+1). Anal. calcd
for C6H9ClFNO2: C, 39.69; H, 5.00; N, 7.71. Found: C, 39.87; H, 5.11; N, 7.45.
Synthesis of 4-methyl-2,4-methanoproline
Methyl 2-(N-(2-methylallyl)benzamido)acrylate (11)
Method A: A solution of KOtBu (3.31 g, 29.3 mmol, 2.1 eq.) in THF (100 ml) was
treated under argon at -78 ºC with a solution of methyl N-benzoyl-3-chloroalaninate 7 (3.30 g,
13.7 mmol, 1.0 eq) in THF (50 ml) followed by addition of 10 (5.50 g, 40.8 mmol, 3.0 eq). The
reaction mixture was allowed to warm to room temperature, stirred for 12 h, and partitioned
25
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 42
1
2
3
4
5
6
7
8
9
between CH2Cl2/H2O (500/100 mL). The organic layer was separated, washed with 10% solution
of citric acid, brine, dried over Na2SO4, and evaporated under reduced pressure. The crude
product was purified by flash column chromatography (pentane/EtOAc = 3/1, Rf = 0.6) to afford
the title compound 11 as a colorless oil (2.70 g, 10.8 mmol, 76% yield). Method B: A three
necked, 2 L round-bottomed flask was fitted with a mechanical stirrer and charged with methyl
pyruvate 12 (102.0 g, 1.00 mol) in 1L of toluene; then 2-methylallylamine 13 (71.0 g, 1.0 mol)
was added and the mixture was stirred for 4 hours at room temperature under argon. After
stirring, the reaction mixture was poured into water and the organic layer was separated, dried
over Na2SO4 and filtered. The filtrate was cooled to −10-0ºC and triethylamine (101.0 g, 1.00
mol) was added followed by benzoyl chloride (112.0 g , 0.80 mol) while the temperature was
maintained. The resulting mixture was allowed to warm to room temperature and stirred
overnight. The precipitate of triethyamine hydrochloride was filtered and washed with toluene.
The filtrate was concentrated under reduced pressure and purified by flash chromatography
(pentane/EtOAc = 3/1, Rf = 0.6) to afford the title compound 11 as a yellow oil (155.4 g, 0.60
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
mol, 60% yield). H NMR (400 MHz; methanol-d4; Me4Si), δ: 1.72 (s, 3H, CH3), 3.51 (s, 3H,
OCH3), 4.24 (s, 2H, CH2), 4.89 (br s, 2H, CH3C=CHH), 5.64 (s, 1H, NC=CHH), 6.01 (s, 1H,
13
NC=CHH), 7.13-7-51 (5H, m, Ph). C NMR (101 MHz, methanol-d4; Me4Si), δ: 20.7 (s), 53.0
(s), 55.9 (broad s), 114.1 (s), 122.9 (broad s), 129.0 (s), 129.3 (s), 131.6 (s), 137.1 (s), 141.7 (s),
142.2 (s), 165.6 (s), 173.4 (broad s). m/z (CI): 260 (M+1). Anal. calcd for C15H17NO3: C, 69.48;
H, 6.61; N, 5.40. Found: C, 69.11; H, 6.32; N, 5.71.
Methyl 2-benzoyl-4-methyl-2-azabicyclo[2.1.1]hexane-1-carboxylate (Bz-6-OMe)
The compound 11 (11.8 g, 0.045 mol) was dissolved in distilled acetonitrile (1 L) and
acetophenone (2.5 g, 0.021 mol) was added. The reaction was irradiated at 366 nm, and the
26
ACS Paragon Plus Environment

Page 27 of 42
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
reaction progress was monitored by NMR. After the reaction completeness, the mixture was
concentrated under reduced pressure and the residue was purified by flash column
chromatography using pentane/EtOAc = 2/1 as an eluent. Rf = 0.4.The product Bz-6-OMe as
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
white solid (11.6 g, 0.042 mol, 94% yield). M. p. = 100-101 ºC. H NMR (400 MHz; CDCl3;
Me4Si), δ: 1.28 (s, 3H), 1.84 (br s, 2H, 2 × CHH), 1.98 (br s, 2H, 2 × CHH), 3.37 (s, 2H, NCH2),
3.79 (s, 3H, CH3), 7.40 (t, 3JH-H = 7.6 Hz, 2H, Ph), 7.51 (t, 3JH-H = 7.3 Hz, 1H, Ph), 7.75 (d, 3JH-H
= 7.0 Hz, 2H, o-Ph). 13C NMR (101 MHz; CDCl3; Me4Si), δ: 16.5 (CH3), 43.7 (s, CCH3), 46.3 (2
× CH2), 52.2 (s, OCH3), 59.8 (s, NCH2) 68.1 (CCO2CH3), 128.3 (s, CH, Ph), 128.5 (s, CH, Ph),
131.4 (s, CH, Ph), 134.4 (s, C, Ph), 169.0 (s,CO2Me), 173.5 (s, NCOPh). m/z (CI): 260 (M+1).
Anal. calcd for C15H17NO3: C, 69.48; H, 6.61; N, 5.40. Found: C, 69.23; H, 6.84; N, 5.12.
1-Carboxy-4-ethyl-2-azabicyclo[2.1.1]hexan-2-ium chloride (6∙HCl)
A suspension of Bz-5-OMe (0.73 mol, 200 g) in 6N HCl (1L) was brought to reflux for
10 hours, then cooled to 0 ºC, the resulting precipitate (benzoic acid) was filtered, and washed
with 200 ml of cold water. The filtrate was concentrated under reduced pressure. The resulting
solid residue was triturated in dry acetone to afford the title product 6∙HCl as a white solid (124.3
g, 0.7 mol, 97% yield). M. p..> 250 ºC. 1H NMR (700 MHz; D2O; Me4Si), δ: 1.26 (s, 3H, CH3),
1.82 (dd, J = 5.9 and 2.4 Hz, 2H, 2×CHH), 2.18 (d, J = 6.0 Hz, 2H, 2× CHH), 3.23 (s, 2H,
13
NCH2). C NMR (176 MHz; D2O; Me4Si), δ: 14.4 (s, CH3), 44.1 (s, 2×CH2), 45.5 (s, CCH3),
52.6 (s, NCH2), 68.9 (s, NCCOOH), 168.9 (s, COOH). m/z (CI): 142 (M+1). Anal. calcd for
C7H12ClNO2: C, 47.33; H, 6.81; N, 7.89. Found: C, 47.68; H, 6.52; N, 7.51.
Synthesis of 2,4-methanoproline
1-Carboxy-2-azabicyclo[2.1.1]hexan-2-ium chloride (4∙HCl)
27
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 28 of 42
1
2
3
4
5
6
7
8
9
N-Boc amino acid Boc-4 (91 mg, 0.4 mmol) was stirred in 3 M hydrogen chloride in
dioxane solution (4 mL) for 14 hours. The solvent was removed under reduced pressure, the
residue was taken up in water (5 mL) and freeze-dried to afford title compound 4∙HCl as a white
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
powder (60 mg, 0.37 mmol, 92%). M. p. > 200 °C. H NMR (700 MHz; D2O; Me4Si), δ: 1.73
(dd, J = 6.0 and 2.2 Hz, 2H, 2×CHH), 2.37 (br m, 2H, 2× CHH), 2.89 (t, J = 3.3 Hz, 1H, CH),
3.40 (s, 2H, NCH2). The 1H NMR spectrum is consistent with the literature reported one (in
methanol-d4).42 13C NMR (176 MHz; D2O; Me4Si), δ: 36.8 (s, CH), 40.2 (s, 2×CH2), 49.4 (s,
NCH2), 71.6 (s, NCCOOH), 169.2 (s, COOH). m/z (CI): 128 (M+1). Anal. calcd for
C6H10ClNO2: C, 44.05; H, 6.16; N, 8.56. Found: C, 44.12; H, 5.99; N, 8.40.
Synthesis of the model compounds
2-Acetyl-2-azabicyclo[2.1.1]hexane-1-carboxylic acid (Ac-4)
4∙HCl (50 mg, 0.31 mmol) was mixed with acetic anhydride (0.8 mL, 8.5 mmol, 22
equiv.) and triethylamine (0.2 mL, 1.4 mmol, 3.7 equiv.) in tetrahydrofuran – dichloromethane
mixture (7 mL / 10 mL). The mixture was stirred at the room temperature until clear solution was
obtained (14 hours). Solvents were removed under reduced pressure, the residue was dissolved in
water (2 mL) in order to quench residual amounts of the anhydride. Resulting solution was
freeze-dried, the residue was taken up in minimal amount of water (0.3 mL) and eluted through a
short cation exchange resin (Dowex 50WX8, 50-100 mesh, volume 2 mL) by deionized water.
Acidic fractions were collected and freeze-dried. Ac-4 was afforded as a white powder (52 mg,
3.1 mmol, quant.). HRMS, m/z (ESI): calculated for C8H12NO3 [M+H]+ 170.0812; found
+
170.0809. 1H NMR (700 MHz; D2O+KDSO4; Me4Si), δ: 3.55 (s, 2H, δ-CH2), 2.78 (t, J = 3.4 Hz,
1H, γ-CH), 2.12 (m, 2H, β-CH), 1.98 (s, 3H, CH3C=O), 1.62 (dd, J = 4.8, 2.0 Hz, 2H, β-CH). 13C
28
ACS Paragon Plus Environment

Page 29 of 42
The Journal of Organic Chemistry
1
2
3
4
5
6
NMR (126 MHz; D2O+KDSO4; Me4Si), δ: 173.4 (s, CO2H), 172.7 (s, O=C-N), 70.1 (s, α-C),
52.2 (s, δ-CH2), 42.1 (s, β-CH2), 34.0 (s, γ-CH), 20.7 (s, CH3C=O).
7
8
9
2-Acetyl-4-fluoro-2-azabicyclo[2.1.1]hexane-1-carboxylic acid (Ac-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
Compound was synthesized in analogous procedure starting from 5∙HCl (59 mg, 0.32
mmol) using dichloromethane as a solvent. The target compound Ac-5 was obtained as a
+
yellowish powder (55 mg, 0.29 mmol, 90% yield). HRMS, m/z (ESI): calculated for C8H11FNO3
1
[M+H]+ 188.0717; found 188.0715. H NMR (700 MHz; D2O+KDSO4; Me4Si), δ: 3.65 (s, 2H,
13
δ-CH2), 2.38 (m, 2H, β-CH), 2.18 (m, 2H, β-CH), 1.94 (s, 3H, CH3C=O). C NMR (126 MHz;
D2O+KDSO4; Me4Si), δ: 173.5 (s, O=C-N), 171.1 (d, J = 11 Hz, CO2H), 86.8 (d, J = 263 Hz, γ-
CF), 60.5 (d, J = 25 Hz, α-C), 50.3 (d, J = 28 Hz, δ-CH2), 47.1 (d, J = 20 Hz, β-CH2), 20.5 (s,
CH3C=O). 19F NMR (659 MHz; D2O+KDSO4; Me4Si), δ: −173.3 (s).
2-Acetyl-4-methyl-2-azabicyclo[2.1.1]hexane-1-carboxylic acid (Ac-6)
6∙HCl (0.3 g, 1.7 mmol) was stirred with acetic anhydride (0.5 mL, 5.3 mmol, 3 equiv.)
and triethylamine (0.4 mL, 2.9 mmol, 1.7 equiv.) in dichloromethane (5 mL) until a clear
solution was obtained (1 hour). The solvent was removed under reduced pressure, the residue
was taken up in water (5 mL) and freeze-dried. Resulting crude material was taken up in water (3
mL) and this solution was passed through a cation exchange resin column (3.5 mL), eluted with
pure water. Acidic fractions were collected and freeze-dried. Ac-6 was obtained as a white
+
powder (0.30 g, 1.6 mmol, 97% yield). HRMS, m/z (ESI): calculated for C9H14NO3 [M+H]+
1
184.0968; found 184.0965. H NMR (700 MHz; D2O+KDSO4; Me4Si), δ: 3.38 (s, 2H, δ-CH2),
1.95 (s, 3H, CH3C=O), 1.89 (d, J = 4.6 Hz, 2H, β-CH), 1.67 (dd, J = 4.7, 1.9 Hz, 2H, β-CH),
1.21 (s, 3H, γ’-CH3). 13C NMR (126 MHz; D2O+KDSO4; Me4Si), δ: 173.0 (s, O=C-N), 172.8 (s,
29
ACS Paragon Plus Environment