The Role of the Amino Protecting Group during Parahydrogenation of Protected Dehydroamino Acids

Article abstract of DOI:10.1021/acs.jpca.5b06802

A series of dehydroamino acids endowed with different protective groups at the amino and carboxylate moieties and with different substituents at the double bond have been reacted with parahydrogen. The observed ParaHydrogen Induced Polarization (PHIP) effects in the ¹H NMR spectra are strongly dependent on the amino protecting group. DFT calculations allowed us to establish a relationship between the structures of the reaction intermediates (whose energies depend on the amido substitution) and the observed PHIP patterns.

Full text of DOI:10.1021/acs.jpca.5b06802

Subscriber access provided by University of Otago Library
Article
The Role of the Amino Protecting Group During Para-
Hydrogenation of Protected Dehydroamino acids
Erika Cerutti, Alessandra Viale, Carlo Nervi, Roberto Gobetto, and Silvio Aime
J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.5b06802 • Publication Date (Web): 28 Oct 2015
Downloaded from http://pubs.acs.org on October 31, 2015
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 Physical Chemistry A 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 Physical Chemistry
1
2
3
4
5
6
7
8
9
The Role of the Amino Protecting Group during
Parahydrogenation of Protected Dehydroamino
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
Erika Cerutti,a Alessandra Viale,* ,a Carlo Nervi,*,b Roberto Gobetto,b Silvio Aimea
Department of Molecular Biotechnologies and Health Sciences, University of Torino, Via Nizza
52 – 10126 Torino – Italy
KEYWORDS: PHIP; DFT; hydrogenation mechanism; amino acids
ABSTRACT
A series of dehydroamino acids endowed with different protective groups at the amino and
carboxylate moieties and with different substituents at the double bond have been reacted with
1
parahydrogen. The observed ParaHydrogen Induced Polarization (PHIP) effects in the H NMR
spectra are strongly dependent on the amino protecting group. DFT calculations allowed to
establish a relationship between the structures of the reaction intermediates (whose energies
depend on the amido substitution) and the observed PHIP patterns.
ACS Paragon Plus Environment
1

The Journal of Physical Chemistry
Page 2 of 42
1
2
3
4
Introduction
5
6
7
8
9
Hyperpolarization NMR methodologies continue to be under intense scrutiny. A major driving
force is represented by the possibility of overcoming the intrinsic low sensitivity of the NMR
technique, as high signal enhancements can be obtained upon the application of
hyperpolarization procedures. Several methods are currently being used to generate
hyperpolarized molecules, including Optical Pumping and Spin Exchange (SEOP) of noble
gases, Dynamic Nuclear Polarization (DNP), ParaHydrogen Induced Polarization (PHIP) and
Signal Amplification By Reversible Exchange (SABRE). 1ꢀ6 From the technologial point of
view, among these modalities PHIP is the less demanding one, as it relies on the hydrogenation
of unsaturated substrates by using hydrogen enriched in its singlet spin state (parahydrogen).
This reaction yields products where both 1H and heteronuclei NMR resonances can be greatly
enhanced.4,5 A number of studies on hyperpolarized molecules have been reported, for the
elucidation of chemical reactions mechanisms (by taking advantage of the detection of
intermediates at very low concentrations),4 for the setꢀup of novel analytical protocols7 and in the
Magnetic Resonance Imaging field.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
In the latter field, the search for hyperpolarized molecules is mainly oriented towards
substrates of metabolic interest. Pyruvate6 and lactate8 generated by the DNP approach, and
succinate9 generated by PHIP are the most relevant hyperpolarized biomolecules. Of course
amino acids are also of great potential interest, and some examples of amino acids
hyperpolarized by the DNP method are reported.10,11
The most straightforward precursors for the preparation of hyperpolarized amino acids by
PHIP are the corresponding dehydroamino acids. The use of proper protecting groups at the
ACS Paragon Plus Environment
2

Page 3 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
amino and carboxylate functionalities is required in order to attain high hydrogenation yields. Up
to now, parahydrogen induced polarization in amino acids has been investigated for elucidating
the reaction mechanism,12 in the case of unsaturated precursors of protected alanine,12 in a small
peptide with antibiotic properties,15 in some neurotransmitters,16,17 in bioactive peptides labelled
with propargylꢀamino acids,18 in ester derivatives of amino acids in water,19 and in some amino
acids and peptides by the SABRE methodology.20
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
Here, the PHIP effects observed in the parahydrogenation of a number of protected amino
acids are reported, establishing a correlation between the observed effects and the reaction
pathway. On the ground of these results it is evidenced that the effects of the protective groups
should be taken into account in the development of a PHIPꢀbased efficient hyperpolarization
protocol for natural amino acids.
Results
1H PHIP experiments
A series of dehydroamino acids (LNa, where N=1ꢀ11) have been reacted with parahydrogen to
afford the corresponding hydrogenated derivatives LNe (chart 1). Protecting groups on both the
carboxylic and the amino groups have been introduced in order to carry out the
parahydrogenation reactions in organic solvents. Removal of the carboxylate protection by
hydrolysis in basic conditions affords waterꢀsoluble substrates which have also been tested for
parahydrogenation in water solution.
ACS Paragon Plus Environment
3

The Journal of Physical Chemistry
Page 4 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 1. Dehydroamino acids for parahydrogenation.
Parahydrogenation reactions have been carried out using ALTADENA conditions21 by using
either [1,4ꢀbis(diphenylphosphino)butane](1,5ꢀcyclooctadiene)rhodium(I) tetrafluoroborate or
[(1,4ꢀbis[(phenylꢀ3ꢀpropanesulfonate)phosphine]butane](norbornadiene)rhodium(I)
tetrafluoroborate as catalysts in deuterated organic solvent (acetone, methanol, chloroform) and
water (D2O), respectively. Single scan 1H spectra have been acquired in order to observe the
PHIP patterns.
10 mM solutions of all the derivatives L1aꢀL11a have been hydrogenated to the corresponding
saturated products L1eꢀL11e in 100% yield after 10 seconds reaction in all the solvents but
chloroform, where the reaction has been found to continue also during spectra acquisition, thus
affording a PASADENA22 pattern in the 1H NMR spectra, with an overall hydrogenation yield of
about 50%.
In acetone, methanol and water, the 1H PHIP pattern is strongly dependent on the nature of the
amino protective group, with no relevant solvent effect. When R3 is a methyl (L2) or a proton
ACS Paragon Plus Environment
4

Page 5 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
(L3, L9), net absorption / emission signals typical of the ALTADENA experiments are observed
for the three protons of the product molecules (Figure 1, spectra b, c, f), while when R3 is a
phenyl (L1, L8) or a tꢀbutanoyl (L4) moiety, each of the H1, H2 and H3 proton signals appears as
an antiphase peak, (figure 1, spectra a, d and e). Similar results are obtained for R1 = phenyl and
R1 = methyl, indicating that the substitution at the amino acid chain does not strongly affect the
PHIP pattern of H1,H2 and H3 protons. Nevertheless, on going for example from L3 to L9 to L10
(R1 = Ph, CH3 and H respectively), signal enhancements become progressively lower, down to
zero in L10e (see table 1). As a consequence of this general trend, polarization which is
transferred to R1, via isotropic mixing and possibly dipolar interaction, is finally detected only in
L3 but not in L9. Being the T1 values for L3, L9 and L10 quite similar (10.0 s, 12.5 s and 10.5 s
respectively for H1, 2.7/2.6 s, 2.3/2.5 s and 3.4 s respectively for H2/H3, see table S1 in
supporting information), the observed reduction in signal enhancement can not be accounted for
by the relaxation properties of the compounds, but rather on the basis of the reaction rates, as it
will be outlined below: slowlyꢀreacting compounds result to be less polarized due to polarization
loss during the reaction pathway.
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 the case of R3 = H (L3, L7, L9) the formyl proton results to be polarized as well (spectra c
and f in figure 1, table 1, figure S3 in supporting information). Polarization can be transferred to
the formyl proton through the network of spinꢀspin couplings and possibly by dipolar interaction.
When R2 = H (L7), the signal enhancement is observed predominantly at the formyl proton (see
figure S3 in supporting information). In this case, as in L5 and L6, the presence of a COOH
group influences the relaxation characteristics of the systems, causing a shortening of the T1
values for H1, H2 and H3 which prevents polarization to be preserved on these positions (see
figures S1 and S2 supporting information).
ACS Paragon Plus Environment
5

The Journal of Physical 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
Figure 1. Single scan 1H NMR spectra of hyperpolarized (a) L1e, (b) L2e, (c) L3e, (d) L4e, (e)
L8e and (f) L9e (10 mM solutions, RT, 9.4 T; acetoneꢀd6). No additional information is present
in the omitted parts of the spectra.
ACS Paragon Plus Environment
6

Page 7 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
Compound
L1e
Ar
1.4
4.8
10.0
2.1
1
H1
5.2
24.5
20.1
9.3
1
H2/H3
7.0/6.3
18.1/18.9
13.0/13.2
10.2/11.9
1
formyl
N/A
N/A
13.4
N/A
N/A
N/A
17.3
N/A
2.3
L2e
L3e
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
L4e
L5e
~1
1
1
L6e
2.3
1
~1
1
L7e
4.9
6.5
1
2.5/2.7
5.4/6.7
1
L8e
N/A
N/A
N/A
L9e
1
L10e
L11e
20.9
23.4/21.2
N/A
Table 1. Proton signal enhancements for compounds Lne, calculated as the ratios between the
integrals in the polarized spectra and in spectra of relaxed products (a value of 1 indicates that
the proton is not polarized).
When R1 = H (L10), no enhanced signals are observed under these experimental conditions
(see figure S4 in supporting information). This is not the case for R1 = H, R3 = CF3 (compound
L11), where 1H net PHIP signals are observed for H1 and the just formed methyl group (figure
S5 in supporting information). Also in this case, relaxation measurements show that this
behaviour can not be accounted for by the T1 values of H1, H2, H3 in L10e and L11e, as they are
quite similar (10.5 s and 12.4 s respectively for H1, 3.4 s and 3.0 s respectively for the just
formed methyl group). We can surmise that the strongly electronꢀattracting CF3 group on the
amidic CO influences the reaction mechanism and/or the lifetime of the involved intermediates
reducing the overall reaction time and thus limiting the polarization loss due to relaxation.
Evidence will be given below.
ACS Paragon Plus Environment
7

The Journal of Physical Chemistry
Page 8 of 42
1
2
3
4
DFT studies of the reaction mechanism
5
6
7
8
9
To get more insight into the determinants of the observed effects, we have computationally
assessed the effect of the protective group R3 on the reaction mechanism. The evaluation of the
energies of intermediates and transition states along the reaction pathway allows to extract
information about the role of R3 in determining the observed PHIP patterns. As no effect of R1
on the pattern of H1, H2 and H3 protons has been experimentally evidenced, the calculations have
been initially performed on a reference substrate, where R1 = R2 = R3 = CH3 (L0a), and then
extended to the set of deyhydroamino acids subjected to the parahydrogenation reactions.
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
The general hydrogenation mechanism (scheme 1), as found from the calculation results,
involves the formation of the ηꢀH2 complexes a (figure 2), which are then transformed into the
dihydride species b. Dihydride intermediates in the parahydrogenation of unsaturated amino
acids have been previously detected by Giernoth et al.12 In the present case, two different
dihydride intermediates may be formed at this stage: in the first one (bC) the equatorial hydride
is on the same side of the β sp2 carbon atom, while in the second one (bN) the equatorial hydride
is on the side of the α sp2 carbon atom. The pathways leading to bC and bN will be called “C
route” and “N route” respectively. In both cases, the b intermediates have very low lifetimes, as
the transfer of the first hydrogen atom to the C=C double bond giving the monohydride
intermediates c (either cC or cN), is very fast for all the substrates. The third reaction step is a
rotation of the monohydrogenated products around the residual RhꢀC bond, giving the rearranged
monohydrides dC and dN respectively. Then, transfer of the second hydrogen atom from the
metal to the substrate occurs, leading to the fully hydrogenated products e.
ACS Paragon Plus Environment
8

Page 9 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
The structures of all the intermediates involved in the reaction pathway and of the
corresponding transition states have been optimized by DFT geometry, and activation energies
for the various reaction steps have been calculated.
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. LNa (N = 0ꢀ11) hydrogenation mechanism. S=solvent; ( = diphenylphosphinobutane
ACS Paragon Plus Environment
9

The Journal of Physical Chemistry
Page 10 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 2. Optimized structure of the η2ꢀH2 Rh complex 0a; nonꢀrelevant hydrogen atoms are
omitted for clarity.
The [1,4ꢀbis(diphenylphosphino)butane](η2ꢀH2)(L0a)Rh+ complex (0a) has a pentacoordinated
trigonal bipyramidal structure, similar to the one previously reported for (DuPHOS)(η2ꢀ
H2)Rh+,23 and [1,4ꢀbis(diphenylphosphinobutane)](η2ꢀH2)Rh+.24,12 In 0a, the remaining axial
coordination site is not occupied by a solvent molecule but by the oxygen atom of the acetamido
moiety (Figure 2), yielding a fiveꢀmembered ring, including the Rh atom. The other oxygen
atoms of L0a may potentially coordinate the metal in place of the acetamido one as well, but in
this case more constrained fourꢀmembered rings would be formed, and this kind of coordination
is thus not energetically favored. Due to the overall asymmetry of the complex, two very similar
structures with a local minimum of energy have been computationally found. In the first one the
ester group is placed in the equatorial plane that includes the Rhꢀalkene moiety on the same side
of phosphorous P1 (Figure 2), while in the second one it is on the opposite side. However, these
two structures have almost the same energy (0.84 kJ/mol of difference) and therefore they can be
considered equivalent.
ACS Paragon Plus Environment
10

Page 11 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
Starting from the DFT optimized structure of the intermediate ηꢀH2 complex 0a, the transition
states (TS) leading to the intermediates 0b (0aCTS and 0aNTS) have been computed.
7
8
9
For the “C route”, the optimized TS structure 0aCTS is only 7.7 kJ/mol higher in energy than 0a
(Figure 3). 0aCTS has only one imaginary frequency at 600i cm–1, and its vibrational mode is
associated with the reaction coordinate that involves motion of the H2 hydrogen atoms towards
the cleavage of the HꢀH bond and the formation of two RhꢀH bonds. All attempts to identify the
0bC intermediate failed. Then it seems that in the “C route” the monohydrogenated complex 0cC
is formed directly from 0a, in an apparent barrierless process, as previously reported.24 Same
results have been found for all the considered ligands. In the monoꢀhydrogenated product 0cC the
newly formed CꢀH bond lies in the Rh equatorial plane and gives rise to an agostic Rh···HꢀC in
plane interaction (2.070 Å), allowing the Rh atom to maintain a pseudoꢀoctahedral coordination
structure (Figure 3). The intermediate product 0cC is 87.7 kJ/mol more stable than 0a. The
energies of the 8cC, 9cC and 11cC intermediates have also been evaluated and are reported in
Table 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
Figure 3. 0aCTS (left) obtained using 0a as starting molecule and the optimized intermediate
product 0cC (right).
ACS Paragon Plus Environment
11

The Journal of Physical Chemistry
Page 12 of 42
1
2
3
4
5
6
7
8
9
#
R1
R2
R3
aTS
10.6
b
bTS
15.1
c
cTS
d
dTS
41.7
e
–
–
14.0
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
–67.8
9N
(551i) 18.8 (636i) 46.5 (51i)
(773i)
–
CH3 CH3
CH3 CH3
CH3 CH3
H
121.9
10.4
(624i)
–
31.6
56.1
(799i)
ꢀ
0
–80.5
–69.4
–85.8
9C
0N
0C
8N
8C
1N
2N
3N
4N
78.8 (41i)
6.4
–
20.1
–
13.4
44.9
(778i)
(550i) 27.0 (584i) 49.6 (39i)
–
CH3
Ph
119.4
7.7
(600i)
–
38.0
56.9
(838i)
ꢀ
0
87.7 (24i)
5.9
–
18.6
–
9.1
51.9
(763i)
–65.9
–82.6
–83.5
(595i) 19.8 (584i) 41.7 (38i)
–
110.0
6.9
(648i)
–
41.2
59.2
(812i)
ꢀ
0
89.0 (24i)
10.4
–
22.6
–
52.6
–
Ph
Ph
Ph
Ph
CH3
CH3
CH3
Ph
CH3
H
(481i) 25.8 (617i) 34.1
(837i) 139.4
11.9 21.0
(472i) 23.7 (637i) 43.0
–
–
34.0
–
–74.8
–81.4
–86.5
–81.2
–93.7
(854i) 139.7
9.2 20.2
(454i) 26.8 (619i) 50.1 (28i)
–
–
10.5
36.2
–
(823i) 150.3
11.4 24.5 13.6
(471i) 28.8 (652i) 42.1 (20i)
–
–
45.7
–
CH3 tꢀBuO
(847i) 155.7
3.3 14.2 12.9
(670i) 27.9 (656i) 70.9 (46i)
–
–
37.1
11
N
(790i)
–
H
CH3
CF3
156.2
8.0
(576i)
–
37.0
62.2
(832i)
11
C
ꢀ
0
93.2 (31i)
Table 2. Energies (in kJ/mol) of the dihydride (b), monoꢀ (c and d) and dihydrogenated (e)
intermediates relative to the starting dihydrogen compounds (a). For transition states barrier
energies relative to the preceding structure (in kJ/mol) and imaginary frequencies (cm–1) are
reported.
ACS Paragon Plus Environment
12

Page 13 of 42
The Journal of Physical Chemistry
1
2
For the “N route”, the 0aNTS transition state is characterized by an energy barrier of 6.4 kJ/mol,
imaginary frequency at 550i cm–1. Its vibrational mode leads to the dihydride 0bN (27.0 kJ/mol
more stable than 0a), that in turn gives the monohydrogenated species 0cN via the TS 0bNTS with
an activation energy of 20.1 kJ/mol and imaginary frequency of 584i cm–1 (Figure 4). The
intermediate product 0cN is 49.6 kJ/mol more stable than 0a, and shows an agostic Rh···HꢀC in
plane interaction (2.363 Å). In both 0cN and 0cC the oxygen of the amidic carbonyl group still
interacts with the metal center in axial position, now affording a five and a sixꢀmembered ring,
respectively. Analogous results have been found for L1, L2, L3, L4, L8, L9 and L11; the
energies of the involved transition states and intermediates are reported in Table 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
The final hydrogenation steps proceed firstly via the rotation around the RhꢀC bond, giving 0d,
which in turn undergoes the final hydrogen transfer from the hydride moiety.
In the “C route”, the substrate in 0cC rotates around the RhꢀC bond bringing the oxygen of the
TS
amido group from the coordination site below the Rh plane to the equatorial plane (via 0cC
,
activation energy of 38.0 kJ/mol and imaginary frequency of 24i cm–1). The resulting species
0dC is slightly less stable than 0cC (1.9 kJ/mol). However, the second hydrogen transfer from
0dC occurs via 0dCTS with an activation energy of 56.9 kJ/mol (838i cm–1), resulting in the final
dihydrogenated product 0e that is 119.4 kJ/mol more stable than the starting 0a. An alternative
pathway, leading directly from 0cC to 0e, without the rotation of the ligand, is also possible, but
in this case a very high energy barrier of 114.1 kJ/mol is found (the TS has a single imaginary
frequency at 902i cm–1, it is 26.4 kJ/mol less stable than 0a, and its vibrational mode is
associated with the reaction coordinate that involves motion of the hydride towards the formation
of the second CꢀH bond). Similar results have been obtained for complexes 9a, 8a and 11a.
ACS Paragon Plus Environment
13

The Journal of Physical Chemistry
Page 14 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 4. 0aNTS (left) obtained using 0a as starting molecule and the optimized intermediate
product 0cN (right).
In the “N route”, the transition state involved in the ligand rotation, 0cNTS, has an activation
energy of 13.4 kJ/mol (imaginary frequency at 39i cm–1), and affords the intermediate 0dN,
which is 19.8 kJ/mol more stable than 0cN. Therefore, the energy required for the rearrangement
of 0c to give 0d is much lower for the “N route” than for the “C route”. 0dN then undergoes the
second hydrogen transfer, via the 0dNTS transition state, characterized by an activation energy of
44.9 kJ/mol (imaginary frequency of 778i cm–1, see Figure 5). The calculated energies for other
ligands (L1, L2, L3, L4, L8, L9, L11) are reported in Table 2.
Figure 5. 0dNTS and the fully hydrogenated product 0e.
ACS Paragon Plus Environment
14

Page 15 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
In 0e, the fully hydrogenated substrate is still weakly coordinated to the Rh metal via the
oxygen on its equatorial plane (2.199 Å) and the agostic H (2.103 Å) interactions, both in the “C
route” and in the “N route”, but only in the latter case the solvent (which has not been taken into
account up to now) should easily replace the hydrogenated product 0e.
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
The overall energy diagram of the hydrogenation reaction is reported in Figure 6.
Figure 6. Schematic energy diagram for the complete hydrogenation reaction, as from
computational results. Red: “C route”; black: “N route”. **: the direct pathway from 0cC to 0e
(no rearrangement) is possible but energetically unfavored.
Discussion
The hydrogenation mechanism involves the formation of four reaction intermediates, from the
ηꢀH2 complex (a) to the dihydride species (b) to the monohydrogenated derivative (c), which
ACS Paragon Plus Environment
15

The Journal of Physical Chemistry
Page 16 of 42
1
2
3
4
successively undergoes the rearrangement process (d) before the second H transfer takes place.
5
6
Two routes are possible depending on the ligand moiety the first H atom is transferred to.
7
8
9
As expected, the ηꢀH2 complexes have very short lifetimes, being quickly transformed into the
corresponding dihydride derivatives. The “N route” appears slightly more favoured than the “C
route”, although the dihydride complexes bC evolve to the monohydrogenated species cC
apparently without activation energy. In the case of the “N route” the involved activation
energies are anyway very low, being between 14.2 and 24.5 kJ/mol. The monohydrogenated
species cC are more stable than the corresponding cN ones. This is because the agostic RhꢀH
interaction is much stronger in cC than in cN. For example, in 0cC and in 0cN the RhꢀH distances
are 2.070 and 2.363 Å, and the just formed CꢀH bonds are 1.135 and 1.092 Å, respectively. For
complexes 9, 0 and 8 the energy required for the rearrangement of c to give d is much lower for
the “N route” (14.0, 13.4 and 9.1 kJ/mol, respectively) than for the “C route” (31.6, 38.0 and
41.2 kJ/mol). For this reason only the “N route” has been considered for calculations involving
the other substrates.
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 indication that the “N route” is more favored with respect to the “C route” comes from
the observation that, whereas in the dC intermediates all the Rh coordination sites are occupied
(with the unsaturated C and the O of the amido moiety on the equatorial plane and the O of the
ester moiety below the plane), in dN, after the rotation, an axial coordination site below the
equatorial plane is available for coordination of a solvent molecule. The quite pronounced
solvent effect on the reaction rate observed when going from acetone/methanol/water to
chloroform may therefore be taken as a further evidence of the predominancy of the “Nꢀroute”.
Other solvents characteristics (such as solvent viscosity or hydrogen solubility) may in principle
affect the reaction rate. Nevertheless the employed solvents at the reaction temperatures used
ACS Paragon Plus Environment
16

Page 17 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
here have very similar viscosities (ranging between 0.3 and 0.55 Pa*s, having acetone at RT and
chloroform at 333K – the best and the worst solvent respectively for what concerns the reaction
rate – viscosity values of 0.306 and 0.39 Pa*s respectively),13 and the higher hydrogen pressure
used for reactions in water should compensate for the lower solubility in this solvent with respect
to the organic ones (the time for reaction completion is actually the same in water, acetone and
methanol).
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
On this basis, the effect of R3 on the H PHIP patterns may be taken as a reporter of the extent
of an asymmetrical relaxation process unbalancing the spin levels populations, occurring during
the hydrogenation reaction.25 Hydrogenation reactions catalyzed by complexes of the type
(diphosphine)(diene)Rh+ are known to proceed via the soꢀcalled “unsaturate route”:26 first, the
unsaturated substrate coordinates the metal center of the catalyst, then molecular H2 enters the
coordination sphere and it is transferred to the substrate. The transfer is very fast, thus allowing
to preserve some polarization up to the final hydrogenated product. The amount of lost
polarization is related to the lifetime of intermediates. The redistribution of the spins among the
energy levels which leads to adsorption/emission 1H signals even in ALTADENA conditions
(when only net signals are expected) can take place in intermediates in which the two
parahydrogen atoms are magnetically non equivalent, if their lifetime is long enough.
As shown by the DFT calculations, some intermediates along the reaction pathway are
stabilized by specific interactions of the metal with the ligand and/or with the solvent, thus
increasing their lifetimes. Among these, b, c and d contain the two parahydrogen atoms in
magnetically non equivalent positions. The effect of R3 on the PHIP pattern may be accounted
for by the interaction of the amido carbonyl group with the metal center, occurring in these three
species, as found by Giernoth et al.,12 leading to a stabilization of the intermediate which is
ACS Paragon Plus Environment
17

The Journal of Physical Chemistry
Page 18 of 42
1
2
3
4
5
6
7
8
9
dependent on the electronic and/or steric properties of R3. The stronger the COꢀRh interaction,
the longer the intermediate lifetime and hence the efficiency of the asymmetrical relaxation
process leading to the change in the spin populations and hence to antiphase 1H signals in
ALTADENA experiments.
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
The short lifetimes of the dihydride species b suggest that these intermediates play a limited
role in determining the observed PHIP patterns. On the other hand, it is reasonable to think that
in c and d the interaction of the amido carbonyl group in axial position with the metal center is
effective in stabilizing these intermediates, as it yields a five/six membered ring. The molecular
rearrangement in c to yield d and the transfer of the second H atom in d to yield e are the slowest
reaction steps. Altogether, these considerations support the view that the intermediates c and/or d
are those in which the redistribution of the spins among the nuclear levels due to asymmetrical
relaxation can take place, to a different extent depending on the strength of the stabilization
and/or on the effect of the axial interaction on the following reactions steps, either before (c
species) or after (d species) the substrate rotation takes place.
As the PASADENAꢀlike pattern is observed when R3=Ph, but not when R3=CH3, H (see figure
1a, 1b and 1c), one may surmise that the lifetimes of c and/or d apparently become longer by
increasing the steric hindrance of R3. One may think that bulkier R3 substituents induce slower
rotation of the ligand during the transfer of the second hydrogen atom to the substrate, thus
preventing the rapid evolution to d, and finally to e. As DFT calculations suggest that the slowest
reaction step is the last one (from d to e), a further important contribution to the observed
behaviour derives from the electronic properties of R3, and in turn from its ability to stabilize
intermediates d by the axial RhꢀO interaction.
ACS Paragon Plus Environment
18

Page 19 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
R1 does not affect the PHIP pattern (net or antiphase signals), but it influences the overall
reaction rate, causing a lowering in the signal enhancements of all the protons on going from R1
= Ph (L3) to R1 = Me (L9) to R1 = H (L10). This trend is in accordance with the computed
activation energies of the last hydrogenation step, which become progressively higher from L3 to
L9 to L10 (table 2). Being the T1 values for the three hydrogenated products quite similar, the
polarization loss is higher for compounds with the longerꢀliving intermediates d.
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
L11 appears as a different case, as the electronꢀwithdrawing properties of the CF3 group in R3
affect the overall reaction mechanism. The transition state 11aTS has a very low energy barrier
(only 3.3 kJ/mol), with the last hydrogenation step being among the lowest energy barriers (37.1
kJ/mol). This allows the polarization to be preserved even in the presence of the strong dipolar
interactions of the three proton resonances of the methyl group in the product.
There are other possible mechanisms leading to antiphase signals in ALTADENA
experiments. Among these, the nonꢀadiabacity of sample transfer from low to high magnetic
field may play a role in the present case due to the use of manual transfer, which may be
subjected to variations from one sample to the other. Nevertheless, the procedure has been
accurately standardized and each sample measurement has been repeated many times in order to
rule out any effect of manual errors.
On the other hand, we can not completely rule out a contribution from the possible
continuation of reactions inside the magnet, during the few seconds interpassing between the
insertion of the sample in the NMR magnet and the spectrum acquisition. This would lead to
antiphase signals for those systems where the intermediates are longerꢀlived and hence the
reaction rate is lower.
ACS Paragon Plus Environment
19

The Journal of Physical Chemistry
Page 20 of 42
1
2
3
4
5
6
7
Conclusions
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
In summary, from the analysis of experimental PHIP data and theoretical calculations on a
number of variously substituted dehydroamino acids, new insights into the hydrogenation
pathway and the effect of the protective groups on the reaction rates and intermediates lifetimes
have been gained. Even though the “C route” cannot completely be ruled out, there are strong
indications that the operating mechanism follows the “N route”, i.e. the first hydrogen transfer
occurs to the α carbon atom of the alkene. A molecular rearrangement in the monohydrogenated
intermediate c to give d occurs, and, together with the second H transfer to yield the final fully
hydrogenated products, it is the slowest reaction step. A strong dependence of the PHIP pattern
on the nature of the amino protecting group has been shown: there is a delicate relationship
between the bulkiness of the substituent, its donating/withdrawing electronic properties and the
lifetime of the monohydrogenated intermediates c and d, in which a redistribution of the spin
populations among the energy levels can occur, thus yielding antiphase patterns of the PHIP
signals in the 1H NMR spectra of the products, even if the reactions are carried out under
ALTADENA conditions. We can not completely rule out that this pattern is due to a lowering of
the reaction rate induced by the different substituents, which would cause the reaction to partially
proceed after introduction of the sample in the magnet.
The information gained in this work should be taken into account when designing unsaturated
precursors of amino acids to be parahydrogenated with the aim of preparing hyperpolarized 13Cꢀ
amino acids, as the redistribution of the spin populations, occurring at some stage of the reaction
ACS Paragon Plus Environment
20

Page 21 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
pathway, and the lowering of the overall reaction rate due to stabilization of intermediates has to
be minimized in order to attain the highest 13C polarization in the products.
7
8
9
The presence of a strong electronꢀattractor amino protecting group, promoting a faster
reaction, yields positive consequences on the observed polarization. It is worth to note that the
trifluoroacetic protection is easy to remove by hydrolysis, which is of importance for the
preparation of unprotected hyperpolarized amino acids by parahydrogenation of unsaturated
precursors followed by hydrolysis of the protective groups.
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 Section
1H and 13C NMR spectra were recorded on a JEOL EXꢀ400 spectrometer operating at 400
MHz for 1H and 100 MHz for 13C respectively. For paraꢀH2 experiments, single scan 1H spectra
were acquired. Signal enhancements were calculated by comparing absolute integrals in the 1H
PHIP spectra and in reference spectra acquired after complete relaxation using the same NMR
parameters. T1 values were measured on degassed samples by the inversion recovery experiment
at 14 T, 298K .
Solvents were stored over molecular sieves and purged with nitrogen before use. Hydrogen
was produced by a CLAIND generator, model HG300.
Paraꢀenriched hydrogen (about 51%) was prepared storing H2 over Fe2O3 at 77 K for one hour.
Parahydrogenation reactions were carried out in ALTADENA21 conditions in a 5 mm NMR
tube equipped with a Young valve. For experiments in organic solvents, [1,4ꢀ
bis(diphenylphosphino)butane](1,5ꢀcyclooctadiene)rhodium(I) tetrafluoroborate (1 mg) was
dissolved in acetoneꢀd6 or CD3OD or CDCl3/acetoneꢀd6 5:1 (0.4 mL), and activated by reaction
ACS Paragon Plus Environment
21

The Journal of Physical Chemistry
Page 22 of 42
1
2
3
4
5
6
7
8
9
with H2. The solution was then frozen and degassed, and the substrate (0.004 mmol) and the
paraꢀH2 enriched mixture (4 atm) were added. For experiments in water, the substrates were
added to 0.4 ml of a 2mM solution of [(1,4ꢀbis[(phenylꢀ3ꢀ
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
propanesulfonate)phosphine]butane](norbornadiene)rhodium(I) tetrafluoroborate in D2O,
prepared according to the published method,8 under Ar atmosphere, and paraꢀH2 (8 atm) was
added. The hydrogenation was carried out by shaking the tube for 10 seconds at RT in acetoneꢀd6
and CD3OD, 333 K in CDCl3/acetoneꢀd6 and 353 K in D2O. All the experiments have been
repeated three times under the same reaction conditions in order to check reproducibility. No
relevant variation was observed.
Hydrogenation yields were evaluated by reacting the substrates with normal hydrogen under
the same experimental conditions used for reactions with parahydrogen and measuring the
amounts of product and (if present) remaining substrate at successive reaction times (5, 10, 15
seconds shaking, and 60 seconds for reactions in chloroform) by 1H NMR. In all cases except for
chloroform solutions, conversion was complete after 10 seconds shaking.
Calculations were performed with the Gaussian 09 (G09) program package27 employing the
DFT method with Becke’s three parameter hybrid functional28 and LeeꢀYangꢀParr’s gradient
corrected correlation functional29 (B3LYP). The LanL2DZ basis set and effective core potential
were used for the Rh atom, and the splitꢀvalence 6ꢀ31G** basis set was applied for all other
atoms. Geometries of the complexes were optimized in the gas phase without any constrain, and
the nature of all stationary points was confirmed by normalꢀmode analysis. Thermal correction
based on harmonic frequencies and Gibbs free energy calculations were performed at 298.15 K
and 1 atm. The nature of transition states were confirmed by harmonic vibrational frequency
ACS Paragon Plus Environment
22

Page 23 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
calculations and normalꢀmode analyses, which gave for each of all transition states herein
presented a single value of imaginary frequency.
7
8
9
Synthetic procedures
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
(Z)ꢀmethyl 2ꢀbenzamidoꢀ3ꢀphenylacrylate (L1), (Z)ꢀmethyl 2ꢀacetamidoꢀ3ꢀphenylacrylate
(L2), (Z)ꢀmethyl 2ꢀformamidoꢀ3ꢀphenylacrylate (L3), (Z)ꢀ2ꢀbenzamidoꢀ3ꢀphenylacrylic acid
(L5), (Z)ꢀ2ꢀacetamidoꢀ3ꢀphenylacrylic acid (L6), and methyl 2ꢀformamidoacrylate (L10) were
prepared according to literature procedures.30,26
(Z)ꢀmethyl 2ꢀ(tertꢀbutoxycarbonylamino)ꢀ3ꢀphenylacrylate (L4)27
0.55 g (2.51 mmol) of diꢀtertꢀbutyl dicarbonate and a catalytic amount of 4ꢀ
(dimethylamino)pyridine (0.03 g, 0.23 mmol) were added to a stirred solution of L3 (0.5 g, 2.28
mmol) in 13 mL of acetonitrile, and the solution was stirred at room temperature for 3 hours.
Acetonitrile was then evaporated and the residue was partitioned between water and diethyl
ether. The aqueous phase was extracted with diethyl ether (2x30 mL). The combined organic
layer was washed with brine (3x30 mL), dried over Na2SO4, filtered and the solvent was
evaporated under reduced pressure. A pale yellow solid of (Z)ꢀmethyl 2ꢀ(Nꢀ(tertꢀ
butoxycarbonyl)acetamido)ꢀ3ꢀphenylacrylate (0.67 g, 92%, 1H NMR (acetoneꢀd6, 400 MHz): δ =
7.66 (1H, s, CH), 7.52 (2H, m, Ar), 7.43 (3H, m, Ar), 3.80 (3H, s, COOCH3), 2.5 (3H, s,
COCH3), 1.29 (9H, s, 3xCH3). 13C NMR (acetoneꢀd6, 100 MHz): δ = 172.6 [C], 165.2 [C], 152.2
[C], 137.4[C], 134.0 [C,], 130.8 [CH], 130.0 [2xCH], 129.8 [2xCH], 128.3 [C], 83.8 [C] 52.7
[CH3], 27.7 [3xCH3], 26.1 [CH3]) was obtained and used with no further purification.
ACS Paragon Plus Environment
23

The Journal of Physical Chemistry
Page 24 of 42
1
2
3
4
5
6
7
8
9
0.46 mL (4.2 mmol) of N,Nꢀdimethylethylene diamine were added to a stirred solution of 0.67
g (2.10 mmol) of (Z)ꢀmethyl 2ꢀ(Nꢀ(tertꢀbutoxycarbonyl)acetamido)ꢀ3ꢀphenylacrilate in 13 mL of
acetonitrile, and the solution was stirred overnight at room temperature. Acetonitrile was then
evaporated and the residue was partitioned between water and diethyl ether. The aqueous phase
was extracted with diethyl ether (2x50 mL). The combined organic phase was washed with brine
(3x50 mL), dried over Na2SO4, filtered and the solvent was evaporated under reduced pressure.
The crude was then purified on silica gel using diethyl ether/petroleum ether 50/50 to afford L4
as a white solid (0.47 g, 67%). 1H NMR (acetoneꢀd6, 400 MHz): δ = 7.68 (2H, d, Ar), 7.41 (3H,
m, Ar), 7.22 (1H, s, CH), 3.79 (3H, s, CH3), 1.41 (9H, s, 3xCH3). 13C NMR (acetoneꢀd6, 100
MHz): δ = 166.8 [C], 154.4 [C], 135.1 [C], 132.0 [CH], 130.7 [2xCH], 129.9 [CH], 129.3
[2xCH], 127.3 [C], 80.2 [C], 52.5 [CH3], 28.4 [3xCH3]
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
(Z)ꢀ2ꢀformamidoꢀ3ꢀphenylacrylic acid (L7)
0.14 g of L3 were dissolved in 5 mL of methanol, and 15 mL of a 0.05 M NaOH in water
were added. The mixture was stirred for 3 hours at room temperature. Methanol was then
evaporated and the remaining water solution was acidified with HCl to pH=2. L7 precipitated as
a white solid and was collected by filtration. (0.100 g, 78%). 1H NMR (acetoneꢀd6, 400 MHz): δ
= 8.21 [1H, s, CHO], 8.74 (1H, bs, NH), 7.82 (2H, d, Ar), 7.38ꢀ7.32 (4H, m, 3xAr, CH). 13Cꢀ
NMR (acetoneꢀd6, 100 MHz): δ = 172.7 [C], 154.3 [C], 134.1 [C], 131.4 [CH], 129.6 [2xCH],
128.8 [CH], 130.2 [2xCH], 126.5 [C].
(Z)ꢀmethyl 2ꢀbenzamidobutꢀ2ꢀenoate (L8)
ACS Paragon Plus Environment
24

Page 25 of 42
The Journal of Physical Chemistry
1
2
2ꢀphenyloxazolꢀ5(4H)ꢀone (1g, 6.20 mmol, prepared according to the literature procedure)28
was dissolved in 10 mL of dichloromethane and 0.68 mL (15.55 mmol) of acetaldehyde and 6.3
g (62 mmol) of Al2O3 were added. The mixture was stirred at room temperature for 24 h, then
the alumina was filtered off and the solvent was evaporated under reduced pressure. The crude
was purified on silica gel using diethyl ether/petroleum ether (40/60) to afford (Z)ꢀ4ꢀethylideneꢀ
2ꢀphenyloxazolꢀ5(4H)ꢀone as a white solid (0.21 g, 20%). 1H NMR (acetoneꢀd6, 400 MHz): δ =
8,07 (2H, d, Ar), 7.68 (1H, t, Ar), 7.59 (2H, t, Ar), 6.73 (1H, q, CH), 2.21 (3H, d, CH3).
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
0.2 mL of NaOMe 37% in methanol were added to a stirred suspension of (Z)ꢀ4ꢀethylideneꢀ2ꢀ
phenyloxazolꢀ5(4H)ꢀone (0.2 g, 1.1 mmol) in 3 mL of methanol. The solution was stirred at
room temperature for 30 minutes and then methanol was evaporated. The residue was partitioned
between a 10% aqueous solution of ammonium chloride and dichloromethane. The aqueous
phase was extracted with dichloromethane (2x20 mL), the combined organic layers were washed
with NH4Cl (2x25 mL) and water (2x25 mL), dried over Na2SO4, filtered and the solvent was
evaporated affording 0.2 g of L8 as a dense colorless oil in 83% yield. 1H NMR (acetoneꢀd6, 400
MHz): δ = 8.91 (1H, bs, NH), 8.00 (2H, d, Ar), 7.56 (1H, m, Ar), 7.48 (2H, m, Ar), 6.75 (1H, q,
CH), 3.69 (3H, s, COOCH3), 1.79 (3H, d, CHCH3). 13C NMR (acetoneꢀd6, 100 MHz): δ = 166.2
[C], 165.5 [C], 135.0 [C], 134.1 [CH], 132.4 [CH], 129.2 [2xCH], 128.4 [2xCH], 127.0 [C], 52.2
[CH3], 14.0 [CH3].
(Z)ꢀmethyl 2ꢀformamidobutꢀ2ꢀenoate (L9)26
0.50 g (2.95 mmol) of LꢀThreonine methyl ester hydrochloride and a catalytic amount of Et3N
(10 uL) were added to a stirred suspension of K2CO3 in methyl formate under argon atmosphere
and the solution was stirred overnight at room temperature. The mixture was then filtered to
ACS Paragon Plus Environment
25

The Journal of Physical Chemistry
Page 26 of 42
1
2
3
4
5
6
7
8
9
remove K2CO3 and methyl formate was evaporated under reduced pressure. The crude was
purified on silica gel with 100% diethyl ether as eluent to afford 0.34 g of L9 as a white solid
(80%). 1H NMR (acetoneꢀd6, 400 MHz): δ = 8.22 (1H, s, CHO), 6.69 (1H, q, CH), 3.71 (3H,
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
COOCH3), 1.74 (3H, d, CHCH3). 13C NMR (acetoneꢀd6, 100 MHz): δ = 164.6 [C], 159.6 [C],
133.5 [CH], 127.2 [C], 52.6 [CH3], 14.4 [CH3]
Methyl 2ꢀ(2,2,2ꢀtrifluoroacetamido)acrylate (L11)
0.80 mL of Triethylamine (5.75 mmol) and 0.55 mL of ethyl trifluoroacetate (4.6 mmol) were
added under Ar atmosphere to a stirred solution of methyl 2ꢀaminoꢀ3ꢀhydroxypropanoate
hydrochloride (0.36 g, 2.3 mmol) in dry methanol. The reaction was quenched after 6 hours with
aqueous HCl 0.1 M and diethyl ether was added. The aqueous phase was extracted with diethyl
ether (2x40 mL) and the combined organic layers were washed with brine (2x40 mL), dried over
Na2SO4, filtered and the solvent was evaporated under reduced pressure to afford a colorless oil
(0.36 g, 72%) of methyl 3ꢀhydroxyꢀ2ꢀ(2,2,2ꢀtrifluoroacetamido) propanoate. 1HꢀNMR (acetoneꢀ
d6, 400 MHz): δ = 8.53 (1H, bs, NH), 4.62 (1H, m, CH), 3.97 (2H, m, CH2), 3.73 (3H, s, CH3).
13CꢀNMR (acetoneꢀd6, 100 MHz): δ = 170.1 [C], 157.8 [C, q, 2JCꢀF = 37.23 Hz], 116.9 [C, q, JCꢀF
= 285.6 Hz], 61.8 [CH2], 56.2 [CH], 52.8 [CH3].
Methyl 3ꢀhydroxyꢀ2ꢀ(2,2,2ꢀtrifluoroacetamido)propanoate (0.35 g, 1.62 mmol) was dissolved
in 10 mL of dry dichloromethane under Ar atmosphere. The solution was cooled at 0°C and 0.16
mL (2.02 mmol) of methanesulfonyl chloride and 0.68 mL (4.86 mmol) of triethylamine were
slowly added; the mixture was then stirred for 3 hours at room temperature. The reaction was
quenched with aqueous HCl 0.1 M. The aqueous phase was extracted with dichloromethane
(2x40 mL) and the combined organic layers were washed with brine (2x40 mL), dried over
ACS Paragon Plus Environment
26

Page 27 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
Na2SO4, filtered and the solvent was evaporated under reduced pressure to afford an orange
liquid (0.100 g, 32%). 1HꢀNMR (acetoneꢀd6, 400 MHz): δ = 9.33 (1H, bs, NH), 8.12 (1H, s,
CHO), 6.11 (2H, s, CH2), 3.85 (3H, s, CH3). 13CꢀNMR (acetoneꢀd6, 100 MHz): δ = 163.8 [C],
156.0 [C, 2JCꢀF = 40.7 Hz], 131.8 [C], 116.4 [C, q, JCꢀF = 287.5 Hz], 114.2 [CH2], 53.5 [CH3].
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
Acknowledgements
We gratefully acknowledge MIUR (PRIN2010, contract no. 2010B5B2NL) and Compagnia
San Paolo (TO_call03_2012_0055) for financial support.
Supporting Information. 1H PHIP spectra of L5e, L6e, L7e, L10e and L11e and T1 values for
L3e, L5e, L6e, L7e, L9e, L10e and L11e are available free of charge via the Internet at
http://pubs.acs.org
AUTHOR INFORMATION
a: Department of Molecular Biotechnologies and Health Sciences, University of Torino, Via
Nizza 52 – 10126 Torino – Italy
b: Department of Chemistry, University of Torino, Via Pietro Giuria 7 – 10125 Torino – Italy
Corresponding Authors
alessandra.viale@unito.it; carlo.nervi@unito.it
ACS Paragon Plus Environment
27

The Journal of Physical Chemistry
Page 28 of 42
1
2
3
4
Author Contributions
5
6
7
8
9
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript.
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
REFERENCES
(1) Abragam, A.; Goldman, M. Principles of Dynamic Nuclear Polarisation, Reports on
Progress in Physics 1978, 41, 395ꢀ467.
(2) ArdenkjærꢀLarsen, J.ꢀH.; Fridlund, B.; Gram, A.; Hansson, G.; Hansson, L.; Lerche,
M.H.; Servin, R.; Thaning, M.; Golman, K. Increase in SignalꢀtoꢀNoise Ratio of > 10,000 Times
in LiquidꢀState NMR, Proc. Natl. Acad. Sc. 2003, 100, 10158ꢀ10163.
(3) Brindle, K.M.; Bohndiek, S.E.; Gallagher, F.A.; Kettunen, M.I. Tumor Imaging using
Hyperpolarized 13C Magnetic Resonance Spectroscopy, Magn. Reson. Med. 2011, 66, 505ꢀ519.
(4) Duckett, S.B.; Sleigh, C.J. Applications of the Parahydrogen Phenomenon: A Chemical
Perspective, Progr. Nucl. Magn. Reson. Spectrosc. 1999, 34, 71ꢀ92.
(5) Green, R.A.; Adams, R.W.; Duckett, S.B.; Mewis, R.E.; Williamson, D.C.; Green,
G.G.R. The Theory and Practice of Hyperpolarization in Magnetic Resonance using
Parahydrogen, Progr. Nucl. Magn. Reson. Spectrosc. 2012, 67, 1ꢀ48.
(6) Viale, A.; Aime, S. Current Concepts on Hyperpolarized Molecules in MRI, Curr. Opin.
Chem. Biol. 2010, 14, 90ꢀ96.
ACS Paragon Plus Environment
28

Page 29 of 42
The Journal of Physical Chemistry
1
2
3
4
5
6
7
8
9
(7) Ellena, S.; Viale, A.; Gobetto, R.; Aime, S. Paraꢀhydrogen Induced Polarization of Siꢀ29
NMR Resonances as a Potentially Useful Tool for Analytical Applications, Magn. Reson Chem.
2012, 50, 529ꢀ533.
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
(8) Chen, A.P.; Kurhanewicz, J.; Bok, R.; Xua, D.; Joun, D.; Zhang, V.; Nelson, S.J.; Hurd,
R.E.; Vigneron, D.B. Feasibility of Using Hyperpolarized [1ꢀ13C]Lactate as a Substrate for in
vivo Metabolic 13C MRSI Studies, Magn. Reson. Imag. 2008, 26, 721ꢀ726.
(9) Chekmenev, E.Y.; Hoevener, J.; Norton, V.A.; Harris, K.; Batchelder, L.S.;
Bhattacharya, P.; Ross, B.D.; Weitekamp, D.P. PASADENA Hyperpolarization of Succinic Acid
for MRI and NMR Spectroscopy, J. Am. Chem. Soc. 2008, 130, 4212ꢀ4213.
(10) Gallagher, F.A.; Kettunen, M.I.; Day, S.E.; Lerche, M.; Brindle, K.M. 13C MR
Spectroscopy Measurements of Glutaminase Activity in Human Hepatocellular Carcinoma Cells
using Hyperpolarized 13CꢀLabeled Glutamine, Magn. Reson. Med. 2008, 60, 253ꢀ257.
(11) Jensen, P.R.; Karlsson, M.; Meier, S.; Duus, J.O.; Lerche, M.H. Hyperpolarized Amino
Acids for In Vivo Assays of Transaminase Activity, Chem. Eur. J. 2009, 15, 10010ꢀ10012.
(12) Giernoth, R.; Einrich, H.; Adams, N.J.; Deeth, R.J.; Bargong J., Brown, J.M., PHIP
Detection of a Transient Rhodium Dihydride Intermediate in the Homogeneous Hydrogenation
of Dehydroamino Acids, J. Am. Chem. Soc. 2000, 49, 12381ꢀ12382.
(13) Dortmund Data bank, Oldenburg, http://www.ddbst.com/en/EED/PCP/VIS_C174.php,
21/10/2015
ACS Paragon Plus Environment
29