Process Development of Biocatalytic Regioselective 5′-O-Levulinylation of 2′-Deoxynucleosides

Article abstract of DOI:10.1021/acs.oprd.5b00152

An enzymatic process was optimized for regioselective levulinylation of the 5′-hydroxyl group in 2′-deoxynucleosides. The results revealed that the nucleobase protecting group influenced the successful outcome of the enzymatic reaction. The enhanced solubility of 4-tert-butylphenoxyacetyl protected 2′-deoxynucleoside played a major role during acylation reaction enabling the starting nucleosides to be the best substrates for Candida antarctica lipase B. Furthermore, we have developed a packed column protocol as a superior alternative to batch process carried-out in a flask, particularly when scale-up is required. The industrial application of this method was demonstrated via the synthesis of 5′-O-levulinylthymidine on a 25 g scale.

Full text of DOI:10.1021/acs.oprd.5b00152

Subscriber access provided by UNIV OF MISSISSIPPI
Full Paper
Process Development of Biocatalytic Regioselective
5’-O-Levulinylation of 2’-Deoxynucleosides
Alejandro Carnero, Yogesh S. Sanghvi, Vicente Gotor, Susana Fernández, and Miguel Ferrero
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.5b00152 • Publication Date (Web): 16 Jun 2015
Downloaded from http://pubs.acs.org on June 19, 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.
Organic Process Research & Development is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 31
Organic Process Research & Development
1
2
3
4
5
6
7
8
Process Development of Biocatalytic
Regioselective 5’ꢀOꢀLevulinylation of 2’ꢀ
Deoxynucleosides
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
Alejandro Carnero,a Yogesh S. Sanghvi,b Vicente Gotor,a Susana Fernándeza,* and Miguel
Ferrero,a,*
aDepartamento de Química Orgánica e Inorgánica, and Instituto Universitario de
Biotecnología de Asturias, Universidad de Oviedo, 33006ꢀOviedo (Asturias), Spain
bRasayan Inc., 2802 Crystal Ridge Road, Encinitas, CA 92024ꢀ6615, USA
mferrero@uniovi.es, fernandezgsusana@uniovi.es
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 31
1
2
3
4
5
Table of Contents Graphic
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 31
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT. An enzymatic process was optimized for regioselective levulinylation of 5’ꢀ
hydroxyl group in 2’ꢀdeoxynucleosides. The results revealed that the nucleobase protecting
group influenced the successful outcome of the enzymatic reaction. Enhanced solubility of
4ꢀtertꢀbutylphenoxyacetyl protected 2’ꢀdeoxynucleoside played a major role during
acylation reaction enabling the starting nucleosides to be the best substrates for Candida
antarctica lipase B. Furthermore, we have developed a packed column protocol as a
superior alternative to batch process carriedꢀout in a flask, particularly when scaleꢀup is
required. The industrial application of this method was demonstrated via the synthesis of 5’ꢀ
Oꢀlevulinylthymidine on 25 g scale.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS: 2’ꢀDeoxynucleosides, Regioselective protection, Lipase, CALꢀB.
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 31
1
2
3
4
Introduction
5
6
7
8
9
Industrial chemists have been trying to achieve sustainable procedures for many years to
satisfy the regulatory concerns.1 Among these, the principles of green chemistry2 provide a
framework for chemists to implement chemical processes that are environmentally friendly
and economically feasible.3 It is recommended that chemists design atom efficient routes,
which incorporates more benign reagents and generate less waste products.
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
Biocatalytic transformations have met some of these attributes particularly in areas such
as fine chemicals and pharmaceuticals.4 There are examples of a variety of industrially
useful processes ranging from the manufacture of small chiral specialty chemicals to the
synthesis of more complex pharmaceutical intermediates.5 Advances in molecular biology,
bioinformatics and protein engineering has made possible to design an enzyme for a specific
process.6 Biocatalysts are made from renewable sources and are capable of accepting a wide
array of substrates, catalyzing reactions with enantioꢀ regioꢀ and chemoselectivities that
contribute to shorter synthetic pathways. Furthermore, biocatalytic processes are less energy
intensive and generate less waste than conventional organic syntheses. An important feature
is the reusability of enzymes when they are immobilized. The immobilized enzymes
maintain high selectivity and stability offering easy of handling during largeꢀscale
operations.
Additionally, continuous flow (CF) processes are emerging as powerful tools to achieve
multiꢀstep synthesis of complex natural products and pharmaceuticals agents under
sustainable conditions.7 CF chemistry results in reduced manual handling, increased process
safety, higher control of the reaction variables, easier reproducibility and reuse of solidꢀ
supported catalyst.8 The CF approach has great synergy when coupled with
4
ACS Paragon Plus Environment

Page 5 of 31
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
biotransformation where enzyme recycling is possible, simplifying reaction workꢀup
between cycles, reduced solvent usage, minimize waste generation and the possibility of
automation. Recent publications on the enzymeꢀcatalyzed processes carried out in CF
systems are the resolution of 2ꢀ(1Hꢀimidazolyl)cycloalkanols,9 1,3,6ꢀtriꢀOꢀbenzylꢀmyoꢀ
inositol,10 2ꢀmethyleneꢀsubstituted cycloalkanols11 or alkaloid calicotomine.12 To the best of
our knowledge, there is only one example of continuous flow enzymatic strategy in the
nucleoside area reported by Gallou et al.13 This work describes a practical synthesis of
isatoribine prodrug via a regioselective hydrolysis catalyzed by Candida antarctica lipase B
(CALꢀB). Furthermore, application of biocatalysis in nucleoside chemistry has been well
recognized as practical alternative to conventional organic synthesis.14
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 our ongoing research related with the preparation of regioselectively protected
nucleoside precursors for oligonucleotide synthesis, we have successfully executed
enzymatic acylations15 and hydrolysis reactions.16 In 2’ꢀdeoxynucleosides, CALꢀB and
Pseudomonas cepacia lipase (PSLꢀC) have demonstrated excellent selectivity towards the
primary hydroxyl group and secondary hydroxyl group, respectively. 5’ꢀOꢀLevulinylated 2’ꢀ
deoxynucleosides were prepared in high yields via acylation reaction catalyzed by CALꢀB
with acetonoxime levulinate or hydrolysis reaction catalyzed by PSLꢀC of the corresponding
dilevulinylated nucleosides. These experiments were originally carried out in a batch mode
and with the advent of CF chemistry, we elected to apply the concepts of CF system for
scaleꢀup using a packed column.
Herein, we describe for the first time a comparison of the enzymatic acylation of 2’ꢀ
deoxynucleosides in batch and packed column mode. The data obtained from batch mode
has been applied to the column process. The key objective is to perform a comparative study
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 31
1
2
3
4
5
6
7
8
9
of several commercial immobilized lipases to first identify the most efficient catalyst and
then establish a scaleꢀup protocol with the preferred lipase. We also evaluated the influence
of the base protecting group on the exocyclic amine of 2’ꢀdeoxynucleosides (dC, dA and
dG) and its impact on the activity and selectivity of the enzyme. It is of paramount
importance to choose the correct protecting groups that ensures their stability during
oligonucleotide synthesis and fast cleavage upon deprotection. Therefore, we studied the
role of conventional base protecting groups (e.g. Bz on dC and dA and Ibu on dG) and
included 4ꢀtertꢀbutylphenoxyacetyl (Tac) as a fast cleaving alternative for nucleoꢀbases.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Results and discussion
The enzymeꢀcatalyzed acylation of 2’ꢀdeoxynucleosides were performed with CALꢀB and
PSLꢀC immobilized on different commercial supports using acetonoxime levulinate as
acylating agent and THF as solvent. We selected four CALꢀB: first Novozym 435, from
Novo Nordisk, adsorbed on Lewait E resin, and three CALꢀB from SPRYNzymes; first
adsorbed on a methacrylic esters matrix, hereinafter CALꢀB1, second adsorbed on
polystyrene resin, hereinafter CALꢀB2, and third covalently immobilized preparation lipase
on epoxy acrylic resin, hereinafter CALꢀB3. With respect to the PSLꢀC, we have observed
excellent results with PSLꢀC from Amano, immobilized on ceramic particles.15h However,
this enzyme is not longer commercially available. It is therefore important to identify an
alternative enzyme. We included three different immobilized PSLꢀC: PSLꢀIM from Amano,
immobilized on diatomaceous earth; adsorbed preparation on polystyrene resin, hereinafter
PSL1; and covalently immobilized preparation on epoxy acrylic resin, hereinafter PSL2.
The last two are from SPRYNzymes.
6
ACS Paragon Plus Environment

Page 7 of 31
Organic Process Research & Development
1
2
3
4
The general enzymatic acylation reaction of 2’ꢀdeoxynucleosides is outlined in Scheme 1.
5
6
Scheme 1
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
First, we studied the acylation of thymidine with various Pseudomonas cepacia lipases in
batch mode. Preliminary enzyme screening was performed at 30 °C and the reaction with 3
equiv of acetonoxime levulinate in THF. This screening exhibited high selectivity toward
the acylation of the secondary hydroxyl group compared to the primary hydroxyl group, for
the three lipases tested. The best conversions were observed with PSLꢀIM and PSL1.
Whereas, the maximal conversion for PSL2 was only 29% after 52 h (entries 1ꢀ3, Table 1).
In order to improve the conversion, higher reaction temperature was evaluated. The results
with PSL1 at 45 °C and 60 °C exhibited similar conversions (69ꢀ75%) compared to 30 °C,
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 31
1
2
3
4
5
6
7
8
9
although with shorter reaction times (entries 2, 5, and 8, Table 1). With PSL2, the
conversion was improved to 60ꢀ61% at higher temperature (entries 6 and 9, Table 1). Best
conversions and selectivity was achieved with PSLꢀIM (entries 4 and 7, Table 1); this lipase
afforded 97% conversion after 19 h at 60 °C and exhibited excellent selectivity during
formation of the 3’ꢀOꢀlevulinyl derivative 2a, only trace amounts of 5’ꢀOꢀacyl and 3’,5’ꢀOꢀ
diacyl products were detected.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Batch mode enzymatic levulinylation of thymidine (1a).
Entry
1
Enzymea
PSLꢀIM
PSL1
T (ºC) t (h)
Conv. (%)b 2a (%)b
3a (%)b
4a (%)b
30
30
30
45
45
45
60
60
60
30
30
30
30
48
52
52
48
32
48
19
26
47
2.5
4
74
74
29
97
75
60
97
69
61
>99
85
89
97c
67
67
23
92
70
53
94
67
58
5
1
6
6
4
4
4
4
2
1
2
4
4
5
5
2
1
3
PSL2
2
4
PSLꢀIM
PSL1
1
5
1
6
PSL2
3
7
PSLꢀIM
PSL1
1
8
1
9
PSL2
1
10
11
12
13
Novozym 435
CALꢀB1
CALꢀB2
CALꢀB3
91
75
78
87
6
6
6
21
5
aSee text or General Experimental Section for detailed description of enzymes.
c
bPercentage of compounds calculated by HPLC from area peaks. After 6 h, the
reaction furnished 85% conversion with 77% formation of 3a.
The reaction of thymidine with CALꢀB was also performed in batch mode under similar
conditions. All enzymes tested demonstrated high conversions and selectivity in short
reaction times at 30 °C. Novozym 435 catalyzed the acylation of the 5’ꢀOꢀposition with a
8
ACS Paragon Plus Environment

Page 9 of 31
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
total conversion in 2.5 h (entry 10, Table 1). CALꢀB1 and CALꢀB2 exhibited reduced
activity with 85ꢀ89% conversion (entries 11 and 12, Table 1). CALꢀB3 showed similar
behavior compared to Novozym 435 but the reaction was relatively slower. After 6 h, the
conversion was 85%, and subsequently increased to 97% in 21 h (entry 13, Table 1). Thus,
Novozym 435 proved to be the most efficient among 4 enzymes tested furnishing 91% of
the desired 5’ꢀOꢀacyl nucleoside 3a.
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
Next, the acylation of dCBz (1b) with Pseudomonas cepacia lipase was studied. As PSLꢀ
IM proved to be the most efficient catalyst with thymidine, levulinylation of 1b was first
investigated with this enzyme at 45 °C and 60 °C. The reaction at higher temperature
resulted in an increase in the biocatalytic activity with 91% conversion after 27 h at 60 °C
while maintaining high selectivity (entry 1 vs entry 2, Table 2). The acylation reaction with
PSL1 and PSL2 were highly selectively furnishing exclusively 3’ꢀOꢀLevꢀdCBz (2b).
Notably, a maximum conversion is reached with both lipases after 24 h, and longer reaction
times led to the hydrolysis of the acylated product (entry 3 and 4, Table 2).
Table 2. Batch mode enzymatic levulinylation of dCBz (1b).
Entry Enzymea
T (ºC) t (h)
Conv. (%)b 2b (%)b
3b (%)b
4b (%)b
1
2
3
4
5
6
7
8
9
PSLꢀIM
45
60
60
60
30
30
30
30
45
24
27
24
24
8
64
91
59
47
96
71
71
90c
98
62
88
59
47
14
11
12
15
9
2
0
2
0
0
5
1
1
3
5
PSLꢀIM
1
PSL1
0
PSL2
0
Novozym 435
CALꢀB1
CALꢀB2
CALꢀB3
Novozym 435
77
59
58
72
84
4
8
20
4
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 31
1
2
aSee text or General Experimental Section for detailed description of enzymes.
3
4
5
6
c
bPercentage of compounds calculated by HPLC from area peaks. After 6 h, the reaction
furnished 71% conversion with 60% formation of 3b.
7
8
9
On the other hand, all the Candida antarctica lipases B evaluation demonstrated modest
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
regioselectivity during acylation of the 5’ꢀhydroxyl group (Table 2). Higher conversion was
obtained with Novozym 435, which catalyzed the acylation reaction of 1b with almost
complete conversion after 8 h at 30 °C. To study the influence of the temperature on the
selectivity of the enzyme, the experiment was conducted at 45 °C. Shorter reaction time was
required to reach 98% conversion, and the percentage of 3b increased from 77% to 84%
(entry 9, Table 2).
To further assess this enzymatic process, nucleosides with purine bases were examined.
N6ꢀBenzoylꢀ2’ꢀdeoxyadenosine (1c) was treated with different enzymatic preparations from
PSL. The enzyme PSLꢀIM exhibited total selectivity at 45 °C and afforded 83% maximum
conversion after 7 h (entry 1, Table 3). An increase in temperature from 45 °C to 60 °C does
not result in an increase in conversion rather decrease in reaction regioselectivity (entry 2,
Table 3). PSL1 and PSL2 also showed excellent results in the levulinylation of the 3’ꢀ
hydroxyl of 1c (entries 3 and 4, Table 3); however, both enzymes are less active and lower
conversions are obtained in comparison with PSLꢀIM.
Table 3. Batch mode enzymatic levulinylation of dABz (1c).
Entry
Enzymea
PSLꢀIM
PSLꢀIM
PSL1
T (ºC)
45
t (h)
7
Conv. (%)b
2c (%)b
83
3c (%)b
4c (%)b
1
2
3
4
5
83
84
31
36
97
0
0
0
0
0
1
60
9
78
6
45
7
31
0
PSL2
45
23
4
35
1
Novozym 435
30
4
92
10
ACS Paragon Plus Environment

Page 11 of 31
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
6
7
8
CALꢀB1
CALꢀB2
CALꢀB3
30
30
30
6
6
2
41
30
68
4
2
4
37
28
64
0
0
0
aSee text or General Experimental Section for detailed description of enzymes. Percentage
b
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
of compounds calculated by HPLC from area peaks.
All the enzymes from CALꢀB demonstrated high selectivity during levulinylation of the
primary hydroxyl group of 1c. These experiments conclude that Novozym 435 resulted in
higher conversion rates (97%) during shorter reaction time (entry 5, Table 3). In contrast,
CALꢀB1 and CALꢀB2 reactions furnished 30ꢀ41% conversion (entries 6 and 7, Table 3).
CALꢀB3 increased the conversion until 68% in only 2 h, but the reaction does not proceed
further with longer reaction times (entry 8, Table 3).
When the acylation was performed on N2ꢀisobutyrylꢀ2’ꢀdeoxyguanosine (1d), the
screening with all PSL was unsatisfactory, probably due to the low solubility of starting
nucleoside in THF even at 60 °C. PSLꢀIM gave rise to 33% conversion with poor
regioselectivity. In the case of PSL1 and PSL2 <5% conversion were observed after long
reaction times.
The change from PSL to CALꢀB showed an improvement during acylation process. The
reaction rate was faster with the CALꢀB than with PSL, despite of the lower solubility of the
nucleoside. The best results were obtained with Novozym 435. It catalyzed selective
levulinylation at the 5’ꢀposition with good conversion at 30 °C. After 4.5 h, a 66%
conversion was observed, and after 22.5 h a 75% conversion was accomplished (entry 1,
Table 4). CALꢀB1, CALꢀB2, and CALꢀB3 are active only at very short reaction times. Thus,
after 2 h the reaction did not proceed beyond 33% conversion (entries 2ꢀ4, Table 4).
Additional experiments with Novozym 435 were carried out with the goal of improving
efficiency. First, the enzyme was added in three portions (entry 5, Table 4) which resulted in
11
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 31
1
2
3
4
5
6
7
8
9
lower conversion rate. Second, the impact of temperature was evaluated. When the
temperature was increased from 30 °C to 45 °C and then to 60 °C, the conversion increased
to 84%, and the selectivity remained excellent (entries 6 and 7, Table 4). Third, the
influence of concentration was studied. To improve the solubility of the starting nucleoside
dilute solutions (0.05 M instead of 0.1 M) were used. Under these conditions, a 17% drop in
the conversion was observed (entry 8, Table 4).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 4. Batch mode levulinylation of dGIbu (1d) with different preparations of CALꢀB.
Entry
Enzymea
T (ºC)
30
t (h)
27
2
Conv. (%)b 2d (%)b
3d (%)b
73
4d (%)b
1
2
3
4
5
6
7
8
Novozym 435
CALꢀB1
75
19
10
33
66
64
84
67
1
1
1
0
1
0
3
1
1
0
0
0
1
0
1
1
30
18
CALꢀB2
30
2
9
CALꢀB3
30
2
33
Novozym 435c
Novozym 435
Novozym 435
Novozym 435d
30
27
9
64
45
64
60
22
22
80
60
65
b
aSee text or General Experimental Section for detailed description of enzymes. Percentage
c
of compounds calculated by HPLC from area peaks. Enzyme was added in three portions.
d0.05 M concentration instead of 0.1 M.
This data suggested that nucleoside solubility plays a major role during enzymatic
acylation process. One possible avenue to increase the solubility of the nucleoside substrate
is to install a more hydrophobic protecting group on the base moiety. Among a variety of
available base protecting groups, we elected the hydrophobic 4ꢀtertꢀbutylphenoxyacetyl
group (Tac) for the following reasons. First, Tac group is more hydrophobic than the
conventional Ibu and Bz protecting groups offering enhanced solubility for the
corresponding amidites. Second, Tac group is cleaved at a significantly faster rate compared
12
ACS Paragon Plus Environment

Page 13 of 31
Organic Process Research & Development
1
2
to Ibu and Bz groups under basic conditions.17a Third, use of Tac group enables synthesis of
fragile oligonucleotides due to shorter deprotection cycle. We synthesized NꢀTacꢀprotected
dC, dA and dG following the literature protocols.17b As expected, the overall solubility of
dCTac (1e), dATac (1f), and dGTac (1g) was better than the conventional dCBz, dABz, and dGIbu
protected nucleosides (Figure 1) in THF. Since PSLꢀIM and Novozym 435 have proven to
be the best catalysts during this study, we selected these two enzymes for the acylation of
Tacꢀprotected nucleosides 1e–g.
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. Influence of the nucleobase protecting group on the solubility at 18 ºC of Nꢀ
protected 2’ꢀdeoxynucleosides.
The levulinylation of dCTac (1e), dATac (1f), and dGTac (1g) with PSLꢀIM did not show
significant improvement with respect to the levulinylation of dCBz, dABz, and dGIbu (entries
1ꢀ3, Table 5). More importantly, excellent regioselectivity was observed with dCTac and
dATac despite of longer reaction time. Similar to dGIbu, the levulination of dGTac with PSLꢀ
IM furnished low conversion rate and non selectivity. Next, during levulinylation of 1e–g
with Novozym 435, both conversions and reaction rates were improved demonstrating the
influence of the increased solubility of the Tacꢀprotected nucleosides (entries 4ꢀ7, Table 5).
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 31
1
2
Novozym 435 catalyzed almost quantitative acylation of dCTac (1e) to 5’ꢀOꢀlevulinyl
protected 3e in 1 h while retaining high selectivity. This is a significant improvement
compared to the acylation of dCBz (Table 2) using same lipase. Next, the acylation of dGTac
nucleoside afforded 98% conversion after 7 h at 45 °C, whereas the acylation reaction of
dGIbu gave a maximum conversion of 84% after 22 h at 60 °C (entry 7, Table 4 vs entry 7,
Table 5). Clearly installation of Tacꢀgroup is beneficial to the outcome of enzyme mediated
acylation process furnishing desired selectivity and excellent yield.
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
Table 5. Batch mode enzymatic levulinylation of dCTac, dATac, and dGTac (1e–g).
Entry Substrate Enzymea
T (ºC)
45
t (h) Conv. (%)b
2 (%)b 3 (%)b 4(%)b
1
2
3
4
5
6
7
PSLꢀIM
24
24
26
1
69
85
20
98
97
89
98
69
84
11
0
0
0
0
0
0
0
0
1
1e
1f
PSLꢀIM
45
1
PSLꢀIM
60
9
1g
1e
1f
Novozym 435
Novozym 435
Novozym 435
Novozym 435
30
98
93
89
96
30
1
4
30
4
0
1g
1g
45
7
1
b
aSee text or General Experimental Section for detailed description of enzymes. Percentage
of compounds calculated by HPLC from area peaks.
Additionally, we investigated the impact of various solvents during acylation of dGTac
using Novozym 435. DMF, acetone, 1,4ꢀdioxane, MeCN, and tertꢀbutylmethyl ether
(TBME) were chosen as alternate solvents and tested at 45 °C (Figure 2). All of the solvents
tested did not improve the conversion compared to reactions performed in THF as solvent.
Despite of high solubility of dGTac in DMF, no product formation was detected. We
speculate that in DMF Novozym 435 does not fold to adopt active conformation needed for
14
ACS Paragon Plus Environment

Page 15 of 31
Organic Process Research & Development
1
2
3
4
5
6
acylation. Use of 10% DMF in THF allowed the reaction to proceed but not as fast as THF
alone.
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. Solvent screening for 5’ꢀOꢀlevulinylation of dGTac with Novozym 435 at 45 °C.
In order to scaleꢀup, we utilized a lipase packed column that allows the possibility of
recycling the enzyme while simplifying the workꢀup. The reaction setꢀup utilized a
circulating electromagnetic pump and a glass column packed with the commercial lipase
(Figure 3B). With the focus on scaleꢀup, it was necessary to understand the impact of water
in THF used as solvent. Therefore, we used anhydrous THF (<100 ppm water), freshly
distilled (over sodium) dry THF and commercial THF (<0.03% water) for our studies and
we found no difference in the rate of conversion or selectivity. Therefore, we utilized less
expensive commercial grade THF for these experiments. Next, levulinylation of thymidine
with Novozym 435 in commercial grade THF was compared in both batch mode and packed
column reaction. These experiments were performed with 1 g of T at 0.1 M concentration,
in a ratio of 1:3 acylating agent and 1:1 of lipase, and a flow rate of 22 mL/min. The HPLC
analysis of the product revealed that the regioselectivity in the batch process was reproduced
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 31
1
2
3
4
5
6
7
8
9
in the packed column process. Quantitative conversion was achieved in both systems with
excellent regioselectivity furnishing 5’ꢀOꢀLevꢀT in 74% isolated yield and 99% purity
(HPLC) without column chromatography. However, in packed column (Figure 3B) the
reaction required longer time for completion compared to batch mode (Figure 3A).
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. A) Setꢀup for batch process. B) Setꢀup for packed column process.
In order to make scaleꢀup more economical, we studied the possible reuse of lipase for
multiple reactions. Interestingly in packed column setꢀup the reuse of lipase is easy and
convenient because the column can be used as is for multiple reactions. Therefore, five
sequential runs were successfully carried out maintaining same conversions and selectivity
in all the cycles without removing the lipase (Table 6). No decrease of activity was observed
with yields oscillating between 72% and 77% offering similar reaction times.
Table 6. Enzyme reuse during levulinylation of T (1a) with Novozym 435 in packed
column mode in 6 h.a
Run Conv. (%)b
1a (%)b
2a (%)b
3a (%)b
90
4a (%)b
Yield (%)c Purity (%)d
1
>99
>99
>99
ꢀ
ꢀ
ꢀ
4
4
5
6
6
5
74
75
77
99
99
98
2e
3e
90
90
16
ACS Paragon Plus Environment

Page 17 of 31
Organic Process Research & Development
1
2
3
4
5
4e
5e
99
97
1
3
6
5
89
87
4
5
75
72
98
99
6
aThe reaction was carried out at 30 °C as 0.1 M solution with 1 g of T in a ratio 1:3 of
7
8
9
b
acylating agent and 1:1 of lipase. Percentage of compounds calculated by HPLC from
c
d
e
area peaks. Isolated yields. Calculated by HPLC from area peaks. Recycled enzyme
from the previous run.
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
Encouraged by these results, the packed column protocol was tried for all 2’ꢀ
deoxynucleosides (1 g scale, Table 7). The reaction on 1g of dCBz (1b) furnished 86% of 5’ꢀ
OꢀLevꢀdCBz after 18 h at 45 °C. In the case of dABz (1c) the reaction stopped after 1 h at
72% conversion and did not progress further (entry 2, Table 7). In order to drive the reaction
to completion 1.5 equiv of acylating agent was added at the beginning of the experiment and
another 1.5 equiv after 1 h of reaction. Under these conditions the product precipitated from
the reaction mixture after 2 h (entry 3, Table 7). Gratifyingly, HPLC of the reaction mixture
showed 94% conversion. The 5’ꢀOꢀLev derivative was easily isolated in 61% yield and 92%
purity by simple filtration of the solid. On the other hand, high yield and excellent
selectivity was observed during the acylation using column approach with Tacꢀprotected
nucleosides 1e–g (entries 4ꢀ6, Table 7). The data in Table 7 clearly indicates that Novozym
435 showed highest conversions offering 5’ꢀregioselectivity in nucleosides carrying NꢀTac
group compared to nucleosides with Nꢀbenzoyl or Nꢀisobutyryl groups. As illustrated in
Figure 4, similar conversions were observed during batch process or during packed column
system. The overall productivity was improved during column experiments given the ease
of reaction workꢀup and ability to recycle enzyme.
Table 7. Levulinylation of dN with Novozym 435 during packed column setꢀup.a
Entry
T
(°C)
t (h)
Conv.
(%)b
1 (%)b 2 (%)b 3 (%)b 4 (%)b Yield
Purity
(%)d
1
(%)c
1
dCBz
45
18
97
3
8
86
3
65
96
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 31
1
2
3
4
5
6
7
8
9
2
dABz 30
dABz 30
dCTac 30
dATac 30
dGTac 45
1ꢀ4
2
72
94
98
97
90
28
6
<4
8
68
85
98
93
88
<1
<1
0
3e
4
61
89
87
74
92
98
93
97
3
2
0
5
1.5
10
3
4
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
6
10
1
1
aThe reaction was carried out in 0.1 M concentration with 1 g of dN in a ratio 1:3 of
b
acylating agent and 1:1 of lipase. Percentage of compounds calculated by HPLC from area
peaks. cIsolated yields. dCalculated by HPLC from area peaks. e1.5 equiv of acylating agent
at the beginning and 1.5 equiv after 1 h of reaction.
Figure 4. Comparison of batch and packed column processes.
To demonstrate the scalability of column protocol, 5’ꢀOꢀLevꢀT was synthesized on 10 g
scale. The process was carried out with a ratio of T:enzyme of 1:0.5 (w/w) instead of 1:1.
Identical results were obtained with half the quantity of enzyme. The reaction was complete
in 6 h and afforded the desired product in 75% yield and 99% purity (HPLC). The excess of
acylating agent was recovered and recycled to minimize the waste. Next, the reaction was
18
ACS Paragon Plus Environment

Page 19 of 31
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
further scaledꢀup starting with 25 g of thymidine furnishing 3a in 94% yield and 97% purity.
Further purification by crystallization afforded the desired product in 70% isolated yield and
99% purity by HPLC.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Conclusions
The successful commercialization of oligonucleotide and nucleosideꢀbased drugs is
expected to result in largeꢀscale demand for nucleoside buildingꢀblocks described in this
study. Given the easy access to efficient synthesis of 5’ꢀOꢀlevulinyl protected nucleosides
accomplished herein, we envision increased usage of these molecules during
oligonucleotide synthesis. We have developed a practical and sustainable process for the
preparation of such molecules based on the regioselective enzymatic acylation carried out
both in batch and in a packed column system. The influence of the nucleobase protecting
group during enzymatic acylation was also studied. To the best of our knowledge this is first
report documenting the role and influence of hydrophobic protecting group during
enzymatic acylation of nucleosides. Thus, protection of base with the hydrophobic 4ꢀtertꢀ
butylphenoxyacetyl group (Tac) furnished excellent yield of the regioselective 5’ꢀprotected
nucleosides in shorter reaction times compared to benzoyl or isobutyryl protecting groups.
We demonstrated the impact of the increased solubility of the starting nucleoside which
translated into successful enzymatic reaction. The process was scaleꢀup to 25 g and the ease
of enzyme reuse was established using a packed column reaction setꢀup built with simple
lab equipment.
19
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 31
1
2
3
4
Experimental section
5
6
7
8
General. Candida antarctica lipase B Novozym 435 (adsorbed on Lewait E resin, 9120
PLU/g) was purchased from Novozymes; Candida antarctica lipase B CALꢀB1 (matrix:
DVD crossꢀlinked polymer based on methacrylic esters, 10855 PLU/g), Candida antarctica
lipase B CALꢀB2 (matrix: polystyrene DVD crossꢀlinked resin, 7320 PLU/g), and Candida
antarctica lipase B CALꢀB3 (matrix: acrylic resin, 2632 PLU/g) were purchased from Sprin
technologies. Pseudomonas cepacia lipase PSLꢀIM (immobilized on diatomaceous earth,
943 U/g) was purchased from Amano Enzyme; Pseudomonas cepacia lipase PSL1 (matrix:
polystyrene DVD crossꢀlinked resin, 229 U/g) and Pseudomonas cepacia lipase PSL2
(matrix: acrylic resin, 159 U/g) were purchased from Sprin technologies. Acetonoxime
levulinate was provided from Sapala Organics Pvt. Ltd. Hyderabad, India. TLC plates were
visualized with 2.5% pꢀanisaldehyde, 3.5% sulfuric acid, and 1% acetic acid in ethanol.
HPLC analysis was carried out with a UV detector at 254 nm, using a Mediterranea Sea
column C18, 5 ꢀm (25 x 0.46 cm) from technokroma or a Spherisorb W 5 ꢀm (25 x 0.46
cm) column. 2a–d, 3a–d, and 4a–d have been previously described.15h,16
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
General procedure for the regioselective enzymatic acylation of 2’-deoxynucleosides
in batch process. A suspension of 1 (0.2 mmol), acetonoxime levulinate (0.6 mmol), and
the enzyme (the ratio 1:Candida antarctica lipase B is 1:1 w/w; the ratio 1:Pseudomonas
cepacia lipase is 1:3 w/w) in anhydrous THF (2 mL) under nitrogen was stirred at 250 rpm
for the time and at the temperature indicated in Tables 1ꢀ7. The reactions were monitored by
HPLC.
Acylation of thymidine (1a): Lichrosorb Si 60 column 10 µm (15 x 0.46 cm), 0.8 mL/min
flow; eluyent 5% MeOH/CH2Cl2. tR (1a)= 12.2; tR (2a)= 4.3; tR (3a)= 5.2; tR (4a)= 3.4 min.
20
ACS Paragon Plus Environment

Page 21 of 31
Organic Process Research & Development
1
2
Acylation of dCBz (1b): Spherisorb W column 5 µm (25 x 0.46 cm), 0.8 mL/min flow;
eluyent 4% MeOH/CH2Cl2 for 12 min and then gradient to reach 12% MeOH/CH2Cl2 at 17
min. tR (1b)= 23.1; tR (2b)= 10.4; tR (3b)= 17.3; tR (4b)= 7.3 min.
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
Acylation of dABz (1c): Mediterránea Sea column C18 5 µm (25 x 0.46 cm), 1 mL/min
flow; eluyent 30% MeCN/H2O for 15 min and then gradient to reach 60% MeCN/H2O at 17
min. tR (1c)= 3.6; tR (2c)= 6.7; tR (3c)= 5.7; tR (4c)= 9.9 min.
Acylation of dGIbu (1d): Mediterránea Sea column C18 5 µm (25 x 0.46 cm), 1 mL/min
flow; eluyent 30% MeCN/H2O for 15 min and then gradient to reach 70% MeCN/H2O at 17
min. tR (1d)= 3.3; tR (2d)= 5.9; tR (3d)= 5.2; tR (4d)= 8.3 min.
Acylation of dCTac (1e): Mediterránea Sea column C18 5 µm (25 x 0.46 cm), 1 mL/min
flow; eluyent 30% MeCN/H2O for 15 min and then gradient to reach 60% MeCN/H2O at 17
min. tR (1e)= 12.6; tR (2e)= 17.1; tR (3e)= 16.2; tR (4e)= 20.1 min.
Acylation of dATac (1f): Mediterránea Sea column C18 5 µm (25 x 0.46 cm), 1 mL/min
flow; eluyent 30% MeCN/H2O for 15 min and then gradient to reach 60% MeCN/H2O at 17
min. tR (1f)= 12.4; tR (2f)= 17.2; tR (3f)= 15.8; tR (4f)= 20.7 min.
Acylation of dGTac (1g): Mediterránea Sea column C18 5 µm (25 x 0.46 cm), 1 mL/min
flow; eluyent 30% MeCN/H2O for 15 min and then gradient to reach 60% MeCN/H2O at 17
min. tR (1g)= 11.7; tR (2g)= 16.1; tR (3g)= 14.9; tR (4g)= 18.5 min.
General procedure for the regioselective enzymatic acylation of 2’-deoxynucleosides
in packed column process (1 g of dN scale). To a 0.1 M solution of 1 (1 g) in THF (used
as purchased without previous treatment, <0.03% water content) was added acetonoxime
levulinate (3 equiv), and the mixture was passed using an electromagnetic pump (22
mL/min) through a 1.8 cm diameter glass column with fritted disc (porosity 0 corresponding
21
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 31
1
2
3
4
5
6
7
8
9
to a nominal pore size 160ꢀ250 ꢀm) charged with 1 g of Novozym 435 until completion (see
Tables 6 and 7 for temperatures and reaction times). Solvent was concentrated under
vacuum and the residue was washed with ether to eliminate the excess of oxime ester, which
can be recaptured from the ether layer. Extraction in CH2Cl2/NaHCO3 (saturated aqueous
solution) is necessary if starting nucleoside is present in the residue. In the acylation of
dCTac, dATac, and dGTac this step did not remove the unreacted nucleoside from the reaction
crude.
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
Protocol for purification of 5’ꢀOꢀLevꢀT (3a): The residue is dissolved in CH2Cl2 and the
solution was added to cold (4 ºC) ether and allows to precipitated overnight in the fridge.
The solid is filtered to afford 3a in 72ꢀ75% yield and 99% pure by HPLC.
Protocol for purification of 5’ꢀOꢀLevꢀdCBz (3b): The residue is recrystallized in MeOH to
give 3b in 65% yield and 96% pure by HPLC (3’ꢀOꢀLev derivative was observed).
Protocol for purification of 5’ꢀOꢀLevꢀdABz (3c): 61% yield and 92% pure by HPLC (3’ꢀOꢀ
Lev derivative was observed). To obtain >99% pure, column chromatography is necessary
(5% MeOH/CH2Cl2).
Protocol for purification of 5’ꢀOꢀLevꢀdCTac (3e): The above general protocol afforded 3e
in 89% yield and 98% pure by HPLC. To obtain >99% purity, column chromatography (5%
iPrOH/CH2Cl2) is necessary to eliminate traces of starting nucleoside.
Protocol for purification of 5’ꢀOꢀLevꢀdATac (3f): The above general protocol afforded 3f in
87% yield and 93% pure by HPLC. To obtain >99% purity, column chromatography (5%
iPrOH/CH2Cl2) is necessary to eliminate traces of starting nucleoside and 3’ꢀOꢀLev
derivative.
22
ACS Paragon Plus Environment

Page 23 of 31
Organic Process Research & Development
1
2
Protocol for purification of 5’ꢀOꢀLevꢀdGTac (3g): The residue is poured in MeOH and the
milky solution was allowed to precipitate. The solid is filtered to afford 3g in 74% yield and
97% pure by HPLC.
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
N4-(4-tert-Butylphenoxyacetyl)-3’-O-levulinyl-2’-deoxycytidine
(2e).
Rf
(5%
1
MeOH/CH2Cl2): 0.44; Mp: 135ꢀ137 ºC; H NMR (DMSOꢀd6, 300 MHz):
δ 1.23 (s, 9H, 3
CH3ꢀtBu), 2.11 (s, 3H, CH3ꢀLev), 2.20 (m, 1H, H2'), 2.40 (m, 1H, H2'), 2.50 (t overlapped
with DMSO, 2H, CH2ꢀLev), 2.74 (t, 2H, CH2ꢀLev, 3JHH 6.4 Hz), 3.62 (m, 2H, H5’), 4.08 (s,
1H, H4’), 4.78 (s, 2H, CH2O), 5.19 (d, 1H, H3’, 3JHH 6.0 Hz), 6.13 (dd, 1H, H1’, 3JHH 7.7 Hz,
3
3JHH 6.0 Hz), 6.83 (d, 2H, CaromH, JHH 8.9 Hz), 7.16 (d, 1H, H5, 3JHH 7.5 Hz), 7.29 (d, 2H,
3
3
CaromH, JHH 8.9 Hz), 8.34 (d, 1H, H6, JHH 7.5 Hz), 11.00 (br s, 1H, NH); 13C NMR
(DMSOꢀd6, 75.5 MHz): δ 27.8 (CH2ꢀLev), 29.5 (CH3ꢀLev), 31.3 (CH3ꢀtBu), 33.8 (CꢀtBu),
37.4 (CH2ꢀLev), 38.2 (C2’), 61.1 (C5’), 66.7 (CH2O), 74.8 (C3’), 85.6 (C4’), 86.4 (C1’), 95.6
(C5), 113.9 (CaromH), 126.2 (CaromH), 143.4 (Cipso), 145.3 (C6), 154.3 (C2), 155.4 (Cipso),
162.0 (C4), 169.4 (NꢀC=O), 172.0 (OꢀC=O), 206.9 (C=O); MS (ESI+, m/z): 516 [(M+H)+,
100%].
N6-(4-tert-Butylphenoxyacetyl)-3’-O-levulinyl-2’-deoxyadenosine
MeOH/CH2Cl2): 0.45; Mp: 67ꢀ69 ºC; 1H NMR (DMSOꢀd6, 300 MHz):
(2f).
Rf
(5%
δ
1.21 (s, 9H, 3CH3ꢀ
tBu), 2.11 (s, 3H, CH3ꢀLev), 2.52 (t, 2H, CH2ꢀLev, 3JHH 6.0 Hz + 1H2’ overlapped), 2.74 (t,
2H, CH2ꢀLev, 3JHH 6.0 Hz), 2.98 (m, 1H, H2'), 3.60 (m, 2H, H5’), 4.07 (m, 1H, H4’), 4.97 (s,
2H, CH2O), 5.17 (br s, 1H, OH), 5.35 (d, 1H, H3’, 3JHH 6.0 Hz), 6.44 (dd, 1H, H1’, JHH 2.2
3
3
3
3
Hz, JHH 6 Hz), 6.85 (d, 2H, CaromH, JHH 8.8 Hz), 7.27 (d, 2H, CaromH, JHH 8.8 Hz), 8.66
13
and 8.70 (2s, 2H, H2 + H8), 10.90 (br s, 1H, NH); C NMR (DMSOꢀd6, 75.5 MHz): δ 27.7
(CH2ꢀLev), 29.6 (CH3ꢀLev), 31.3 (CH3ꢀtBu), 33.8 (CꢀtBu), 36.6 (C2’), 37.5 (CH2ꢀLev), 61.5
23
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 31
1
2
3
4
5
6
7
8
9
(C5’), 67.3 (CH2O), 75.0 (C3’), 83.9 (C1’), 85.4 (C4’), 114.1 (CaromH), 123.4 (C5’), 126.1
(CaromH), 142.9 (C8), 143.3 (Cipso), 149.1 (C4), 151.5 (C6), 151.7 (C2), 155.6 (Cipso), 167.5
(NꢀC=O), 172.0 (OꢀC=O), 207.0 (C=O); MS (ESI+, m/z): 541 [(M+H)+, 100%].
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
N2-(4-tert-Butylphenoxyacetyl)-3’-O-levulinyl-2’-deoxyguanosine (2g). Rf (10%
1
MeOH/CH2Cl2): 0.62; Mp: 117ꢀ119 ºC; H NMR (DMSOꢀd6, 300 MHz):
δ 1.23 (s, 9H,
3CH3ꢀtBu), 2.11 (s, 3H, CH3ꢀLev), 2.47 (m, 1H, H2'), 2.50 (t overlapped with DMSO, 2H,
CH2ꢀLev), 2.77 (t, 2H, CH2ꢀLev, 3JHH 6.3 Hz), 2.86 (m, 1H, H2'), 3.58 (m, 2H, H5’), 4.02 (m,
1H, H4’), 4.81 (s, 2H, CH2O), 5.11 (t, 1H, OH, 3JHH 5.3 Hz), 5.29 (d, 1H, H3’, 3JHH 5.5 Hz),
6.21 (dd, 1H, H1’, 3JHH 5.7 Hz, 3JHH 8.8 Hz), 6.87 (d, 2H, CaromH, 3JHH 8.8 Hz), 7.29 (d, 2H,
3
13
CaromH, JHH 8.8 Hz), 8.27 (s, 1H, H8), 11.81 (br s, 2H, NH); C NMR (DMSOꢀd6, 75.5
MHz): δ 27.7 (CH2ꢀLev), 29.6 (CH3ꢀLev), 31.3 (CH3ꢀtBu), 33.8 (CꢀtBu), 36.6 (C2'), 37.4
(CH2ꢀLev), 61.4 (C5'), 66.3 (CH2O), 75.0 (C3'), 83.0 (C1'), 85.2 (C4'), 114.0 (CaromH), 120.4
(C5), 126.2 (CaromH), 137.6 (C8), 143.6 (Cipso), 147.4 (C2), 148.4 (C4), 154.9 (C6), 155.4
(Cipso), 171.2 (NꢀC=O), 172.0 (OꢀC=O), 207.0 (C=O); MS (ESI+, m/z): 556 [(M+H)+,
100%].
N4-(4-tert-Butylphenoxyacetyl)-5’-O-levulinyl-2’-deoxycytidine
MeOH/CH2Cl2): 0.38; Mp: 88ꢀ90 ºC; 1H NMR (DMSOꢀd6, 300 MHz):
(3e).
Rf
(5%
δ
1.23 (s, 9H, 3CH3ꢀ
tBu), 2.08 (m, 4H, H2’ + CH3ꢀLev), 2.31 (m, 1H, H2'), 2.46 (t overlapped with DMSO, 2H,
3
CH2ꢀLev), 2.70 (t, 2H, CH2ꢀLev, JHH 6.3 Hz), 4.03 (m, 1H, H4’), 4.20 (m, 3H, H3’ + 2H5’),
4.78 (s, 2H, CH2O), 5.44 (br s, 1H, OH), 6.12 (t, 1H, H1’, 3JHH 6.3 Hz), 6.83 (d, 2H, CaromH,
3
3
3JHH 8.8 Hz), 7.16 (d, 1H, H5, JHH 7.4 Hz), 7.28 (d, 2H, CaromH, JHH 8.9 Hz), 8.11 (d, 1H,
H6, 3JHH 7.4 Hz), 10.98 (br s, 1H, NH); 13C NMR (DMSOꢀd6, 75.5 MHz): δ 27.5 (CH2ꢀLev),
29.5 (CH3ꢀLev), 31.3 (CH3ꢀtBu), 33.8 (CꢀtBu), 37.4 (CH2ꢀLev), 39.3 (C2'), 63.9 (C5'), 66.6
24
ACS Paragon Plus Environment

Page 25 of 31
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(CH2O), 70.1 (C3'), 84.5 (C4'), 86.4 (C1'), 95.6 (C5) 113.9 (CaromH), 126.1 (CaromH), 143.7
(Cipso), 145.1 (C6), 154.2 (C2), 155.4 (Cipso), 161.9 (C4), 169.3 (NꢀC=O), 172.2 (OꢀC=O),
206.8 (C=O); MS (ESI+, m/z): 516 [(M+H)+, 100%].
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
N6-(4-tert-Butylphenoxyacetyl)-5’-O-levulinyl-2’-deoxyadenosine
MeOH/CH2Cl2): 0.35; Mp: 64ꢀ66 ºC; 1H NMR (DMSOꢀd6, 300 MHz):
(3f).
Rf
(5%
δ
1.21 (s, 9H, 3CH3ꢀ
tBu), 2.04 (s, 3H, CH3ꢀLev), 2.34 (m, 1H, H2'), 2.40 (m, 2H, CH2ꢀLev), 2.64 (t, 2H, CH2ꢀ
3
2
3
Lev, JHH 6.8 Hz), 2.86 (dt, 1H, H2', JHH 13.04 Hz, JHH 6.4 Hz), 3.99 (m, 1H, H4’), 4.12
(dd, 1H, H5’, 2JHH 11.7 Hz, 3JHH 6.0 Hz), 4.25 (dd, 1H, H5’, 2JHH 11.7 Hz, 3JHH 4.1 Hz), 4.48
3
(m, 1H, H3’), 4.96 (s, 2H, CH2O), 5.51 (br s, 1H, OH), 6.44 (t, 1H, H1’, JHH 6.5 Hz), 6.85
3
3
(d, 2H, CaromH, JHH 8.8 Hz), 7.27 (d, 2H, CaromH, JHH 8.9 Hz), 8.63 and 8.65 (2s, 2H,
13
H2+H8), 10.88 (br s, 1H, NH); C NMR (DMSOꢀd6, 75.5 MHz): δ 27.5 (CH2ꢀLev), 29.5
(CH3ꢀLev), 31.3 (CH3ꢀtBu), 33.8 (CꢀtBu), 37.4 (CH2ꢀLev), 38.5 (C2'), 64.0 (C5'), 67.3
(CH2O), 70.5 (C3'), 83.6 (C1'), 84.3 (C4'), 114.0 (CaromH), 123.5 (C5), 126.1 (CaromH), 143.1
(C8), 143.3 (Cipso), 149.0 (C4), 151.4 (C6), 151.7 (C2), 155.6 (Cipso), 167.5 (NꢀC=O), 172.2
(OꢀC=O), 206.8 (C=O); MS (ESI+, m/z): 541 [(M+H)+, 100%].
N2-(4-tert-Butylphenoxyacetyl)-5’-O-levulinyl-2’-deoxyguanosine (3g). Rf (10%
MeOH/CH2Cl2): 0.52; Mp: 88ꢀ91 ºC; 1H NMR (DMSOꢀd6, 300 MHz):
δ 1.21 (s, 9H, 3CH3ꢀ
tBu), 2.06 (s, 3H, CH3ꢀLev), 2.35 (m, 1H, H2'), 2.46 (t overlapped with DMSO, 2H, CH2ꢀ
Lev), 2.66 (t, 2H, CH2ꢀLev, 3JHH 6.3 Hz + 1H2’ overlapped), 3.96 (m, 1H, H4’), 4.10 (dd, 1H,
2
3
2
3
H5’, JHH 11.8 Hz, JHH 5.9 Hz), 4.21 (dd, 1H, H5’, JHH 11.8 Hz, JHH 4.1 Hz), 4.41 (br s,
1H, H3’), 4.81 (s, 2H, CH2O), 5.47 (d, 1H, OH, 3JHH 3.4 Hz), 6.21 (t, 1H, H1’, 3JHH 6.6 Hz),
6.87 (d, 2H, CaromH, 3JHH 8.8 Hz), 7.28 (d, 2H, CaromH, 3JHH 8.8 Hz), 8.18 (s, 1H, H8), 11.77
13
(br s, 2H, NH); C NMR (DMSOꢀd6, 75.5 MHz): δ 27.5 (CH2ꢀLev), 29.6 (CH3ꢀLev), 31.3
25
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 31
1
2
(CH3ꢀtBu), 33.8 (CꢀtBu), 37.4 (CH2ꢀLev), 38.9 (C2'), 64.0 (C5'), 66.4 (CH2O), 70.4 (C3'), 83.0
(C1’), 84.2 (C4'), 114.0 (CaromH), 120.6 (C5), 126.2 (CaromH), 137.8 (C8), 143.6 (Cipso), 147.3
C6), 148.2 (C4), 154.9 (C2), 155.4 (Cipso), 171.1 (NꢀC=O), 172.2 (OꢀC=O), 206.8 (C=O); MS
(ESI+, m/z): 556 [(M+H)+, 100%].
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
N4-(4-tert-Butylphenoxyacetyl)-3’,5’-di-O-levulinyl-2’-deoxycytidine (4e). Rf (5%
MeOH/CH2Cl2): 0.54; Mp: 54ꢀ56 ºC; 1H NMR (DMSOꢀd6, 300 MHz):
tBu), 2.06 (s, 3H, CH3ꢀLev), 2.09 (s, 3H, CH3ꢀLev), 2.31 (dt, 1H, H2’, 2JHH 14.3 Hz, 3JHH 6.8
δ 1.22 (s, 9H, 3CH3ꢀ
2
3
Hz), 2.48 (m, 5H, H2’ + 2CH2ꢀLev), 2.72 (dt, 4H, 2CH2ꢀLev, JHH 12.3 Hz, JHH 6.4 Hz),
4.26 (m, 3H, H4’+2H5’), 4.78 (s, 2H, CH2O), 5.18 (d, 1H, H3’, 3JHH 6.0 Hz), 6.12 (t, 1H, H1’,
2
3
3JHH 6.8 Hz), 6.82 (d, 2H, CaromH, JHH 8.7 Hz), 7.19 (d, 1H, H5, JHH 7.4 Hz), 7.29 (d, 2H,
3
3
CaromH, JHH 8.7 Hz), 8.12 (d, 1H, H6, JHH 7.4 Hz), 11.01 (br s, 1H, NH); 13C NMR
(DMSOꢀd6, 75.5 MHz): δ 27.5 (CH2ꢀLev), 27.7 (CH2ꢀLev), 29.5 (CH3ꢀLev), 31.3 (CH3ꢀ
tBu), 33.8 (CꢀtBu), 37.4 (CH2ꢀLev + C2'), 63.7 (C5'), 66.7 (CH2O), 74.3 (C3'), 82.2, (C4'),
86.8 (C1'), 95.8 (C5), 113.9 (CaromH), 126.1 (CaromH), 143.4 (Cipso), 145.2 (C6), 154.2 (C2),
155.4 (Cipso), 162.1 (C4), 169.3 (NꢀC=O), 172.0 (OꢀC=O), 172.2 (OꢀC=O), 206.9 (C=O);
MS (ESI+, m/z): 614 [(M+H)+, 100%].
N6-(4-tert-Butylphenoxyacetyl)-3’,5’-di-O-levulinyl-2’-deoxyadenosine (4f). Rf (5%
MeOH/CH2Cl2): 0.52; Mp: 61ꢀ63 ºC; 1H NMR (DMSOꢀd6, 300 MHz):
tBu), 2.04 (s, 3H, CH3ꢀLev), 2.10 (s, 3H, CH3ꢀLev), 2.43 (m, 2H, CH2ꢀLev), 2.51 (t, 2H,
δ 1.21 (s, 9H, 3CH3ꢀ
3
3
CH2ꢀLev, JHH 6.4 Hz), 2.57 (m, 1H, H2’), 2.65 (t, 2H, CH2ꢀLev, JHH 6.8 Hz), 2.74 (t, 2H,
CH2ꢀLev, 3JHH 6.4 Hz), 3.14 (m, 1H, H2'), 4.23 (m, 3H, H4’+2H5’), 4.96 (s, 2H, CH2O), 5.38
(m, 1H, H3’), 6.45 (dd, 1H, H1’, 3JHH 7.9 Hz, 3JHH 6.2 Hz), 6.85 (d, 2H, CaromH, 3JHH 8.8 Hz),
7.27 (d, 2H, CaromH, 3JHH 8.9 Hz), 8.66 and 8.67 (2s, 2H, H2+H8), 10.91 (br s, 1H, NH); 13C
26
ACS Paragon Plus Environment

Page 27 of 31
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
NMR (DMSOꢀd6, 75.5 MHz): δ 27.5 (CH2ꢀLev), 27.7 (CH2ꢀLev), 29.5 (CH3ꢀLev), 29.6
(CH3ꢀLev), 31.3 (CH3ꢀtBu), 33.8 (CꢀtBu), 35.4 (C2'), 37.4 (CH2ꢀLev), 37.5 (CH2ꢀLev), 63.6
(C5'), 67.3 (CH2O), 74.3 (C3'), 81.8 (C4'), 83.8 (C1'), 114.1 (CaromH), 123.5 (C5), 126.1
(CaromH), 143.1 (C8), 143.3 (Cipso), 149.1 (C4), 151.5 (C6), 151.8 (C2), 155.6 (Cipso), 167.5
(NꢀC=O), 172.05 (OꢀC=O), 172.11 (OꢀC=O), 206.8 (C=O), 206.9 (C=O); MS (ESI+, m/z):
639 [(M+H)+, 100%].
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
N2-(4-tert-Butylphenoxyacetyl)-3’,5’-di-O-levulinyl-2’-deoxyguanosine (4g). Rf (10%
MeOH/CH2Cl2): 0.75; Mp: 75ꢀ77 ºC; 1H NMR (DMSOꢀd6, 300 MHz):
tBu), 2.08 (s, 3H, MeꢀLev), 2.10 (s, 3H, CH3ꢀLev), 2.50 (m overlapped with DMSO, 5H,
δ 1.25 (s, 9H, 3CH3ꢀ
3
3
H2’+2CH2ꢀLev), 2.70 (t, 2H, CH2ꢀLev, JHH 6.4 Hz), 2.77 (t, 2H, CH2ꢀLev, JHH 6.4 Hz),
3
2.99 (m, 1H, H2'), 4.22 (m, 3H, H4’+2H5’), 4.83 (s, 2H, CH2O), 5.30 (d, 1H, H3’, JHH 5.6
Hz), 6.25 (dd, 1H, H1’, 3JHH 8.5 Hz, 3JHH 5.9 Hz), 6.90 (d, 2H, CaromH, 3JHH 8.8 Hz), 7.30 (d,
2H, CaromH, 3JHH 8.8 Hz), 8.25 (s, 1H, H8), 11.82 (br s, 2H, NH); 13C NMR (DMSOꢀd6, 75.5
MHz): δ 27.5 (CH2ꢀLev), 27.7 (CH2ꢀLev), 29.5 (CH3ꢀLev), 29.6 (CH3ꢀLev), 31.3 (CH3ꢀtBu),
33.8 (CꢀtBu), 35.6 (C2'), 37.37 (CH2ꢀLev), 37.43 (CH2ꢀLev), 63.6 (C5'), 66.3 (CH2O), 74.5
(C3'), 81.7 (C4'), 83.1 (C1'), 114.0 (CaromH), 120.7 (C5), 126.2 (CaromH), 137.7 (C8), 143.6
(Cipso), 147.4 (C2), 148.4 (C4), 154.9 (C6), 155.4 (Cipso), 171.1 (NꢀC=O), 172.1 (OꢀC=O),
172.2 (OꢀC=O), 206.8 (C=O), 207.0 (C=O); MS (ESI+, m/z): 654 [(M+H)+, 100%].
Enzymatic acylation of 25 g of thymidine in packed column process. A solution of
thymidine (25 g, 103.3 mmol) and acetonoxime levulinate (53.3 g, 310 mmol) in THF (1 L,
used as purchased without previous treatment, <0.03% water content) was passed using an
electromagnetic pump (22 mL/min) through a 4.3 cm diameter glass column with fritted
disc (porosity 0 corresponding to a nominal pore size 160ꢀ250 ꢀm) charged with 12 g of
27
ACS Paragon Plus Environment

Organic Process Research & Development
Page 28 of 31
1
2
3
4
5
6
7
8
9
Novozym 435 during 7 h. Solvent was concentrated under vacuum and the residue was
washed with ether to eliminate the excess of oxime ester, which can be recaptured from the
ether layer (at this stage the 5’ꢀOꢀLevꢀT is obtained in 94% yield and 97% purity by HPLC).
The residue is dissolved in CH2Cl2 (125 mL) and the solution was added to cold (4 ºC) ether
(375 mL), and allows to precipitated overnight in the fridge. The solid is filtered to afford
5’ꢀOꢀLevꢀT in 70% yield and 99% pure by HPLC.
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
Acknowledgment. Financial support of this work by the Spanish Ministerio de Ciencia e
Innovación (MICINN) (Projects CTQ2011ꢀ24237 and CTQ2014ꢀ55015) is gratefully
acknowledged.
1
13
Supporting Information Available. Figures giving H, C, and DEPT NMR spectral
data. This information is available free of charge via the Internet at http://pubs.acs.org/.
References
1. Sustainable Industrial Chemistry: Principles, Tools and Industrial Examples. Centi, G.;
Perathoner, S.; Trifiró, F.; Eds; Wiley, October 2009.
2. Anastas, P. T.; Warner, J. C. Green Chemistry: Theory and Practice; Oxford University
Press: New York, 1998, p 30.
28
ACS Paragon Plus Environment

Page 29 of 31
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
3. Bryan, M. C.; Dillo, B.; Hamann, L. G.; Hughes, G. J.; Kopach, M. E.; Peterson, E. A.;
Pourashraf, M.; Raheem, I.; Richardson, P.; Richter, D.; Sneddon, H. F. J. Med. Chem.
2013, 56, 6007ꢀ6021.
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
4. Meyer, H.ꢀP.; Eichhorn, E.; Hanlon, S.; Lütz, S.; Schürmann, M.; Wohlgemuth, R.;
Coppolecchia, R. Catal. Sci. Technol. 2013, 3, 29ꢀ40.
5. Nestl, B. M.; Nebel, B. A.; Hauer, B. Curr. Opin. Chem. Biol. 2011, 15, 187ꢀ193.
6. a) Bornscheuer, U. T. Synlett 2013, 24, 150ꢀ156. b) Bornscheuer, U. T. Huisman, G.
W.; Kazlauskas, R. J.; Lutz, S.; Moore, J. C.; Robins, K. Nature 2012, 485, 185ꢀ194. c)
Behrens, G. A.; Hummel, A.; Padhi, S. K.; Schätzle, S.; Bornscheuer, U. T. Adv. Synth.
Catal. 2011, 353, 2191–2215.
7. a) Wiles, C.; Watts, P. Green Chem. 2014, 16, 55ꢀ62. b) Newman, S. G.; Jensen, K. F.
Green Chem. 2013, 15, 1456ꢀ1472. c) Ley, S. V. Chem. Rec. 2012, 12, 378ꢀ390. d) Wegner,
J.; Ceylan, S.; Kirschning, A. Chem. Commun. 2011, 47, 4583ꢀ4592.
8. a) Puglisi, A.; Benaglia, M.; Chiroli, V. Green Chem. 2013, 15, 1790ꢀ1813. b)
McQuade, D. T.; Seeberger, P. H. J. Org. Chem. 2013, 78, 6384ꢀ6389. c) MaletꢀSanz, L.;
Susanne, F. J. Med. Chem. 2012, 55, 4062ꢀ4098.
9. Porcar, R.; Sans, V.; RíosꢀLombardía, N.; GotorꢀFernández, V.; Gotor, V.; Burguete,
M. I.; GarcíaꢀVerdugo, E.; Luis, S. V. ACS Catal. 2012, 2, 1976ꢀ1983.
10. Manoel, E. A.; Pais, K. C.; Flores, M. C.; Miranda, L. S. M.; Coelho, M. A. Z.; Simas,
A. B. C.; Freire, D. M. G.; de Souza, M. A. R. O. J. Mol. Catal. B: Enzym., 2013, 87, 139ꢀ
143.
29
ACS Paragon Plus Environment