Selective preparation of tetrasubstituted fluoroalkenes by fluorine-directed oxetane ring-opening reactions

Article abstract of DOI:10.3762/BJOC.16.160

The selective ring-opening reaction of fluoroalkylidene-oxetanes was directed by the presence of the fluorine atom, enabling a two-step access to tetrasubstituted fluoroalkenes with excellent geometry control. Despite its small van der Waals radii electronic, rather than steric influences of the fluorine atom governed the ring-opening reaction with bromide ions, even in the presence of bulky substituents.

Full text of DOI:10.3762/BJOC.16.160

Selective preparation of tetrasubstituted fluoroalkenes by
fluorine-directed oxetane ring-opening reactions
Clément Q. Fontenelle, Thibault Thierry, Romain Laporte, Emmanuel Pfund
and Thierry Lequeux*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2020, 16, 1936–1946.
Normandie Université, Laboratoire de Chimie Moléculaire et
Thioorganique LCMT UMR 6507, ENSICAEN, UNICAEN, CNRS,
6 Bd. du Maréchal Juin, 14050 Caen, France
doi:10.3762/bjoc.16.160
Received: 27 May 2020
Accepted: 23 July 2020
Published: 07 August 2020
Email:
Thierry Lequeux* - Thierry.Lequeux@ensicaen.fr
This article is part of the thematic issue "Organo-fluorine chemistry V".
Guest Editor: D. O'Hagan
* Corresponding author
Keywords:
acyclonucleotide; fluorine; monofluoroalkene; oxetane; selective
ring-opening reaction; tetrasubstituted alkene
© 2020 Fontenelle et al.; licensee Beilstein-Institut.
License and terms: see end of document.
Abstract
The selective ring-opening reaction of fluoroalkylidene-oxetanes was directed by the presence of the fluorine atom, enabling a two-
step access to tetrasubstituted fluoroalkenes with excellent geometry control. Despite its small van der Waals radii electronic, rather
than steric influences of the fluorine atom governed the ring-opening reaction with bromide ions, even in the presence of bulky sub-
stituents.
Introduction
The introduction of fluorine atoms into organic compounds is Other main structural modifications of ACN relied on the intro-
known to modify their biological and physiological properties duction of a hydroxy group into the aliphatic chain to improve
and can enhance the half-life of drugs in vivo [1-4]. During the hydrogen bonding with enzymes [10], or of a carbon–carbon
last decade, fluorinated nucleoside analogues have received in- double bond to constrain the aliphatic chain and to limit confor-
creasing interest, as is illustrated by the two pharmaceutical mational changes [11-13]. For the latter, nucleoside analogues
leads gemcitabine (I) and sofosbuvir (II), potent anticancer or (Figure 2, VI) containing a trans-butenyl moiety where the
antiviral agents, respectively (Figure 1) [5,6]. The field of endocyclic C–O bond was replaced by a C=C bond are recog-
acyclonucleotides (ACN) has been explored less, however, the nized by kinases as dUMP surrogate (V) [11]. However, there is
introduction of fluorine atoms showed remarkable effects. The no existing data for the corresponding fluoroalkene (VII), as the
most representative examples are phosphate analogues such as latter was not yet synthetized. It is expected that the introduc-
the nucleoside phosphorylase inhibitor III and acyclic tion of fluorine into the carbon–carbon double bond, in a posi-
nucleotides such as the antiviral agent FPMPA (IV) [7-9]. tion equivalent to the ring oxygen of the naturally occurring
1936

Beilstein J. Org. Chem. 2020, 16, 1936–1946.
Figure 1: Representative fluorinated nucleos(t)ides and acyclonucleotides.
Figure 2: Acyclonucleotides as nucleotide surrogates.
nucleotide, will improve molecular recognition and activity. In alkenes with good control of their geometry [18-21]. In order to
addition, the polarity of the nucleotide and hydrogen-bond develop a selective synthesis for tetrasubstituted fluoroalkenes
accepting capacity with proteins or enzymes would be restored we envisioned an alternative approach starting from fluo-
[14].
roalkylidene-oxetane derivatives and to the end we have studied
the selectivity of the oxetane ring-opening reaction (Figure 3).
The synthesis of fluoroalkene precursors of modified acyclonu-
cleosides (VIII) has been explored by Choi, and more recently Results and Discussion
by us [15-17]. Nevertheless, it was reported that no antiviral ac- The preparation of a series of fluoroalkylidene-oxetanes 1–3
tivity for compounds of series VIII was observed due to the was previously reported from 3-oxetanone through an olefina-
difficulty of phosphorylation of the substrate by kinases [16]. tion reaction with benzothiazoyl sulfones (Scheme 1) [22]. With
The first kinase phosphorylation step is generally rate limiting, these fluoroalkylidene-oxetanes in hands, we studied the selec-
and the prior introduction of a phosphate or phosphonate func- tivity of ring-opening reactions with heteroatom nucleophiles in
tion can circumvent this problem. The preparation of diols VIII order to access tetrasubstituted fluoroalkenes. A control of the
was realized by olefination of a protected 1,3-dihydroxy- geometry of these reactions would allow ready access to novel
propanone (Figure 3). However, the selective introduction of fluorinated ACN precursors.
functional groups is not possible in these diols as the two
hydroxy groups present similar chemical reactivity. Other ap- At the outset the opening of the oxetane ring of 1 by a range of
proaches are available for a selective preparation of mono- nucleophiles was trialed under acidic conditions. Inspired by
fluoroalkenes including olefination or defluorination reactions Yadav et al. [23], methanol (20 equiv) was used as nucleophile
or a sigmatropic rearrangement, but these approaches are in the presence of camphorsulfonic acid (CSA, 1 equiv) in
limited and do not allow the synthesis of tetrasubstituted fluoro- dichloromethane. The reaction was slow and required heating
Figure 3: Olefination approaches and ring-opening of oxetane derivatives.
1937

Beilstein J. Org. Chem. 2020, 16, 1936–1946.
Scheme 1: Preparation of fluoroakylidene-oxetanes and their ring-opening reactions.
(50–65 °C) for 26 h to reach a 75% conversion and afforded a cesium fluoride were unsuccessfully tested. Finally, using the
mixture of E/Z-1a in 60% yield. However, a low E/Z selectivity conditions developed by Burkhard and Carreira [26], the
(40:60) was observed (Table 1, entry 1). When using neat opening of the fluoroalkylidene-oxetane ring was investigated
benzylic alcohol, completion was achieved after 20 h at 80 °C with hydrobromic acid (HBr 33 wt % in AcOH, 2.3 equiv) in
but again no substantial selectivity could be observed for prod- diethyl ether (Table 1, entry 4). This reaction proved faster and
uct 1b (Table 1, entry 2). Inspired by the work of Müller and reached completion after 45 min at 20 °C, giving product 1d
Wang [24,25], substitution by the more nucleophilic 2-mercap- with an excellent yield of 94% and an E/Z selectivity of 89:11.
tobenzothiazol (BTSH, 1.4 equiv) was also possible in the pres- The isomers could be separated and crystals of the major isomer
ence of CSA (1 equiv) at 20 °C and an 80% conversion was were obtained by recrystallization. The X-ray diffraction analy-
reached after 24 h. An improved E/Z ratio of 25:75 was deter- sis clearly showed that the bromine atom was located on the
mined for this reaction and compound 1c was isolated in 65% carbon trans to the fluorine atom resulting in the major product
yield (Table 1, entry 3). In this case the nucleophilic ring- with the E-geometry (see Supporting Information File 1). It
opening reaction appeared to be controlled by steric repulsions should be noted that longer reaction times resulted in slow
between the bulky benzothiazolyl and phthalimidoyl substitu- solvolysis of alcohol 1d with acetic acid giving mainly the cor-
ents affording preferentially the Z-isomer of 1c. However, responding acetate (not shown). Product 1d was also obtained
heteronucleophiles such as sodium azide, secondary amine and in a similar E/Z ratio (88:12) when the ring-opening reaction
Table 1: Oxetane opening by various nucleophiles.
entry
oxetane
Nu
(equiv)
additive
(equiv)
t (h)
T (° C)
solvent
product
E/Z ratioa Yield (%)b
1
2
3
4
5
6
7
8
1
1
1
1
1
2
3
3
MeOH
(20)
CSA
(1)
26
40
CH2Cl2
neat
1a
1b
1c
1d
1d
2d
–
40:60
45:55
25:75
89:11
88:12
94:6
–
60
n.a.
65
94
71
53
–
BnOH
(24)
CSA
(1)
20
80
BTSH
(1.4)
CSA
(1)
24
20
CH2Cl2
Et2O
HBr
(2.3)
AcOHc
0.75
2
20
TBAB
(2.5)
BF3·Et2O
(1.1)
−20
20
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
HBr
(2.3)
AcOHc
0.75
0.75
2
HBr
(2.3)
AcOHc
20
TBAB
(2.5)
BF3·Et2O
(1.1)
−20
3d
96:4
76
aDetermined by 19F NMR of the crude mixture; byield of isolated product; cHBr 33 wt % in AcOH solution.
1938

Beilstein J. Org. Chem. 2020, 16, 1936–1946.
was performed in dichloromethane with tetrabutylammonium study was initiated using non-fluorinated alkylidene oxetanes as
bromide (TBAB) as the bromide source and boron trifluoride substrates. Since the oxetane ring was attacked from the side of
diethyl etherate as an activator (Table 1, entry 5). Nevertheless, the bulky phthalimide or alkyl chain, it appeared plausible that
the isolated yield of the product decreased to 71%. Next, the the selectivity observed for the reaction did not originate from
reaction performed with HBr/AcOH was extended to alkyl- steric hindrance. Accordingly, the electronic influence of fluo-
idene oxetanes substituted by an alkyl chain, and a pyrimidine rine was explored. Thus, the non-fluorinated analogues bearing
base. The presence of the alkyl chain in place of the phthal- a phthalimido group (4) and an alkyl chain (5) were synthe-
imido group did not affect the selectivity observed with 1. The sized, and submitted to the ring-opening reaction [27,28]. Pleas-
ring opening reaction of the n-octyl substituted oxetane 2 ingly, when subjecting compound 4 to the previous reaction
resulted in an excellent selectivity of 94:6 towards the E-isomer conditions, a single isomer of bromoalcohol 4d formed (deter-
of bromoalkylated product 2d (Table 1, entry 6). The E/Z mix- mined by 1H NMR of the crude) and the product was isolated in
ture of 2d was isolated in moderate yield (53%). In this case the an excellent yield of 94% (Table 2, entry 3). The geometry of
corresponding acetate was observed as a minor product (15%) the product was confirmed to be E-4d by X-ray diffraction anal-
but with a similar selectivity of 92:8, although it could not be ysis (see Supporting Information File 1). The attack of the bro-
isolated in pure form. Unfortunately, the introduction of a mide ion this time occurred from the side of the alkene hydro-
nucleic base such as N3-benzoyliodouracil instead of the gen atom (cis attack) and away from the bulky phthalimido
phthalimido group gave a complex mixture of products group giving solely product E-4d. Likewise, the reaction of the
(Table 1, entry 7). In contrast, the ring-opening reaction was alkyl-substituted substrate 5 proceeded with excellent E-selec-
successful when performed in the presence of TBAB and tivity (as determined by 1H NMR of the crude). As in case of
BF3·Et2O and afforded the E-alkene product 3d with good the fluorinated analogue, a substantial part of the newly formed
selectivity (E/Z ratio > 96:4) and 76% yield (Table 1, entry 8). alcohol 5d reacted with acetic acid forming the corresponding
The geometric assignment of compound 3d was corroborated acetate 5d’. Interestingly, it appeared that only the E-isomer of
by 1D NOESY experiments in which after selective irradiation 5d reacted with AcOH giving pure E-5d’ in 44% isolated yield.
of the protons α to the nitrogen atom, a response was observed This left an 81:19 E/Z mixture of 5d from which the pure
only for the protons α to the bromine atom, thus indicating their alcohol E-5d could be isolated in 34% yield (Table 2, entry 4).
spatial proximity.
The overall E/Z ratio for the ring-opening reaction of 5 was
91:9. 1D NOESY experiments were performed with alcohol 5d
In order to elucidate the selectivity control in this ring-opening and the outcome indicated that the bromide attack took place
reaction displayed by bromide ion despite the presence of bulky away from the bulky alkyl chain, resulting in the observed
substituents (phthalimido and alkyl groups), a comparative E-selectivity. These outcomes were opposite to those obtained
Table 2: Selectivity in the presence or absence of the fluorine atom.
entry
substrate
R2
R1
R3
product
E/Z ratioa
yield (%)b
1
2
3
4
1
F
F
H
H
phthalimido
C7H15
H
1d
2d
4d
5d
5d’
89:11
94:6
94
53
94
34d
44
2
H
4c
5
phthalimido
C8H17
H
100:0
81:19
100:0
H
Ac
aDetermined by 19F NMR (1, 2) or 1H NMR (4, 5) of the crude mixture; byield of isolated product; csubstrate 4 was contaminated with 1 molar equiva-
lent of phthalimide; dyield of isolated E-5d.
1939

Beilstein J. Org. Chem. 2020, 16, 1936–1946.
with the fluorinated substrate series, consistent with steric With alkene 8 in hand, the ring-opening reaction was explored
hindrance governing the non-fluorine containing oxetane reac- in the presence of hydrobromic acid (HBr 33 wt % in AcOH) in
tions but electronics influencing and reversing the regioselectiv- diethyl ether (Table 3). As observed with the phthalimido
ity of the fluoro-oxetane reactions.
group, the reaction led to alkene E-9 as the major product,
together with alkene Z-9 and a third product that was identified
The reaction was then extended to fluoroalkylidene-oxetane 8 to as the 2,5-dihydrofuran derivative 10.
expand the range of tetrasubstituted fluoroalkenes accessible via
this method (Scheme 2). A variety of conditions were explored Using the standardized conditions, but at a temperature of 0 °C
to prepare the protected alcohol 8 including an unsuccessful instead of 20 °C, for 30 min, complete conversion was achieved
reduction of the corresponding ethyl ester (vide infra, 12). Then and the three products E,Z-9 and 10 were present in a 62:8:26
we turned our attention to the modified Julia reaction since the ratio as determined by 19F NMR, the remaining 4% being attri-
reduction of the ester functionality could be achieved at the buted to acetylated analogues of 9 (Table 3, entry 1). After
sulfide stage [29], prior to its oxidation to give 6. Alcohol 6 was purification, two products were obtained as a 95:5 mixture and
not stable in basic medium, as a Smiles rearrangement occurred identified by NMR as the desired bromoalcohols E-9 and Z-9,
leading to fluoroethylene and benzothiazolone. Therefore, its respectively. The selectivity of the oxetane ring opening (crude
benzylation was explored under acidic conditions with benzyl E/Z ratio: 89:11) was again governed by the presence of the
trichloroacetimidate (1.5 equiv) and a catalytic amount of tri- fluorine atom and not by steric hindrance.
fluoromethanesulfonic acid. This gave benzyl ether 7 as a 2.5:1
mixture with N-benzylbenzothiazolone (not shown, Scheme 2). In order to limit the competitive formation of the heterocyclic
After purification, benzyl ether 7 was successfully subjected to ether 10, the addition of TBAB as a bromide source was
the modified Julia olefination conditions with 3-oxetanone, to explored. To our delight, after 30 min at 0 °C, the crude
give the corresponding alkene 8 in 79% yield.
19F NMR showed that only 4% of 10 and 96% of 9 as an 84:16
Scheme 2: Synthesis of benzyloxy-substituted fluoroethylidene-oxetane derivative 8.
Table 3: Opening of the benzyloxy-substituted fluoroethylideneoxetane derivative 8.
Entry Nu (equiv)
Additive
(equiv)
Solvent
Reaction
time (h)
Selectivitya
Yield of 9 (%)
E-9
Z-9
10
1
2
HBr (2.3)
AcOH
AcOH
Et2O
0.5
0.5
62
81
8
26
4
45
HBr (1.2)
TBAB (2.5)
CH2Cl2
15
74b
3
TBAB (2.5)
BF3·OEt2 (1.1) CH2Cl2
1
90
10
0
66
aDetermined by 19F NMR of the crude mixture; byield for the 92:8 E/Z mixture.
1940

Beilstein J. Org. Chem. 2020, 16, 1936–1946.
E/Z mixture had formed (Table 3, entry 2). Separation of the Given the previous results, in the presence of the ester function
two E/Z isomers proved challenging by column chromatogra- we expected the ring-opening reaction to proceed with the for-
phy and a 92:8 mixture of E/Z-9 was obtained in 74% yield. mation of additional products to alkenes E-13 and Z-13. In fact,
Finally, the reaction performed with TBAB in the presence of two byproducts formed and were identified as β-hydroxy-
BF3·Et2O afforded only alkenes 9 with an excellent E-selec- methyl-α-fluorolactone 14 and β-bromomethyl-α-fluorolactone
tivity and in 66% yield (Table 3, entry 3). In this case, we 15 (Table 4) [30]. This ring expansion has been already re-
presume 10 was not obtained because the benzyl ether is not ported in the literature from oxetane-containing α,β-unsaturated
nucleophilic enough to react due to it complexation by carbonyl derivatives through a Lewis acid-catalyzed rearrange-
BF3·OEt2.
ment [31]. When the previous conditions (HBr/AcOH) were
tried (3.5 h, 0 °C to 20 °C), hydroxymethyllactone 14 (87%,
To understand in more detail the competitive formation of Table 4, entry 1) was the main product with traces of the corre-
linear versus cyclic products in the reaction, methanol was sponding acetate (4%, not shown), the bromomethyllactone 15
explored as the nucleophile instead of bromide ion (Scheme 3). (8%), and alkene Z-13 (1%). Only lactone 14 could be isolated
The reaction (43 h, 20 °C) realized in MeOH as a solvent and in in a pure form (58% yield). The amount of HBr/AcOH had little
the presence of CSA (1 equiv) afforded a mixture of 10 (85%) influence on the selectivity (Table 4, entry 2) of the reaction.
and the expected methoxyalcohols 11 (10%, 1:1 E/Z ratio) in However, in the presence of TBAB (2.5 equiv) and CSA
addition to starting fluoroalkylidene-oxetane 8 (5%). After (2 equiv) an 84% conversion into mainly brominated lactone 15
purification, compound 10 was isolated in 75% yield.
(79%) with traces of hydroxylated lactone 14, and of
bromoalkene Z-13 was determined by crude 19F NMR. The
remaining 8% appeared to be the β-chloromethyllactone 19 (see
Scheme 7 below), an analogue of 15 as determined by NMR
analysis and supported by HRMS. All three halogenated prod-
ucts were purified and isolated as a mixture (≈61% determined
by NMR). The contrasting result observed with HBr/AcOH and
TBAB/CSA highlighted the importance of the acidity of the
medium on the reaction course of the ring-opening reaction
(Scheme 5). The former, using an excess of acid (33% HBr in
AcOH solution) favored the direct nucleophilic attack (path b)
leading to lactone 14, whereas the latter in the presence of
Scheme 3: Effect of the medium on the selective formation of deriva-
tive 10.
The results from the acid-catalyzed results support the reaction TBAB/CSA allowed bromide addition on the same side of the
outcomes which depended on the nucleophile (methanol or bro- fluorine atom (path a) leading to lactone 15 [31]. However, the
mide ion) (Scheme 4). In the presence of excess bromide ions, reaction can be stopped at the alcohol stage when performed in
the direct intermolecular nucleophilic attack of the oxetane the presence of BF3·Et2O instead of CSA to afford alkene Z-13
(path a) is preferred leading to bromoalcohol 9 (path a). On the as the major product. The use of boron trifluoride etherate as an
other hand, in the presence of the weaker methanol nucleophile, activator combined with TBAB afforded exclusively alkene
an intramolecular ring-opening reaction by the benzyl ether Z-13 as evidenced by TLC, but after work-up, 22% of the
oxygen is preferred leading to the 2,5–dihydrofuran 10 (path b). bromomethyllactone 15 was observed (Table 4, entry 4). It
appeared obvious that Z-13 could cyclize under acidic condi-
Finally, we turned our attention to a last series of reactions tions and during the purification gave β-bromomethyllactone
exploring the ring-opening reaction of fluoroalkylidene-oxetane 15. This was later confirmed when various ester/lactone mix-
12 (Table 4).
tures obtained after the ring-opening reaction were treated with
Scheme 4: Mechanism for the formation of dihydrofuran 10.
1941

Beilstein J. Org. Chem. 2020, 16, 1936–1946.
Table 4: Ring-opening reaction from acetate derivative 12.
entry Nu (equiv)
additive
solvent
t (h)
T (°C)
selectivitya
product
(yield)b
E-13 14
Z-13
15
1
2
3
4
HBr (2.5)
HBr (1.2)
TBAB (2.5)
TBAB (2)
AcOH
Et2O
3.5
3.5
30
4
20
0
–
3
–
–
87
76
7
1
8
14 (58)
–
AcOH
Et2O
7
10
CSA (2)c
BF3.OEt2 (1.5)
CH2Cl2
CH2Cl2
20
–20
6
79 (8)d
22
15 (61)e
–
78
Z-13 (72)f
15 (21)f
5
6
TBAB (1.5)
–
–
CH2Cl2
CH2Cl2
16
2
20
20
–
–
–
–
–
–
–
–
BF3·OEt2 (1)
100
14 (67)
aDetermined by 19F NMR of the crude mixture; bisolated yield; cthe 2nd equivalent of CSA was added after 24 h; dchlorinated instead of brominated
products; eyield of an 87:8:5 mixture of products 15/19/Z-13; fZ-13 and 15 were obtained as a crude mixture, yields were calculated from the crude
mass and 1H NMR.
acid (PTSA) in Et2O giving pure lactone 15 or at least enriched tronic repulsion induced between fluorine and bromine limited
mixtures depending on the reaction times (Scheme 5).
the intermolecular ring-opening reaction by bromide in favor of
a faster intramolecular reaction involving the ester group
As a control, 12 was shown to be unreactive to TBAB on its leading to 14. Indeed, a competitive cyclization reaction
own (Table 4, entry 5). Once again, the effect of the fluorine occurred forming 14 with HBr/AcOH and confirmed when the
atom was highlighted by an investigation with the non-fluori- reaction was performed in the presence of BF3·Et2O only
nated alkylidene oxetane 16. The latter was subjected to the
HBr/AcOH ring-opening reaction conditions (Scheme 6).
Remarkably, the cyclic products β-hydroxymethyl and
β-bromomethyl-γ-lactones 17 and 18 were obtained in an 8:92
ratio [32]. This complete reversal of selectivity in comparison
with fluoroalkylidene-oxetane 12, where the β-hydroxymethyl-
γ-lactone 14 was obtained, confirmed an electronic influence of
the fluorine atom on these ring-opening reactions (Table 3,
entries 1 and 2). In the case of fluoroalkylidene-oxetane 12 and,
Scheme 6: Opening reaction of ethyl 2-(oxetanyl-3-idene)acetate (16).
in contrast with 16, when subjected to HBr/AcOH, the elec-
Scheme 5: Mechanism for the formation of unsaturated lactones 14 and 15.
1942

Beilstein J. Org. Chem. 2020, 16, 1936–1946.
(Table 4, entries 1, 2, and 6). Of note in contrast to 8, when route was not investigated further, and instead functionalization
TBAB/BF3·Et2O was added we cannot exclude a control by of alkene E-1d was explored.
steric or electronic repulsions between bromide ions and the
ester function leading to Z-13 and 15 instead of expected alkene Direct ring-opening reactions of fluoroalkylidene-oxetane 1
E-13 and to the lactone 18 from alkylidene oxetanes 12 and 16, with heteronucleophiles were previously explored and success
respectively.
was only possible with thiols, such as mercaptobenzothiazole
(Table 1). However, the reaction was not selective and afforded
Having established selective approaches for the preparation of a mixture of E/Z alkenes 1c. The functionalization of alkene
halogenated allylic fluoroalkenes, their use in the synthesis of E-1d via displacement of the bromine atom (Scheme 8), with
highly functionalized tetrasubstituted fluoroalkenes was nucleophiles such as CsF and NaN3 was then studied. When the
explored. The derivatization of the three brominated products, reaction was performed in DMF products E-23 and Z-24 were
lactone 15 and alkenes E-1d and E-9, was studied either on the generated in 92% and 93% yield, respectively. Reactions with
bromomethyl (CH2Br) or on the hydroxymethyl (CH2OH) arm, amines and thiols such as pyrrolidine and 2-mercaptobenzothia-
when applicable.
zole, gave rise to the products Z-25 and E-1c in 96% yield, re-
spectively. These reactions were carried out in dichloro-
First, from a mixture of lactones 15 and 19 substitution on the methane in the presence of Et3N. It should be noted that
bromomethyl arm was performed using sodium azide following this two-step method, pure E-1c could be obtained
(Scheme 7). The reaction proceeded smoothly in DMF but it while direct ring opening of 1 with BTSH and CSA resulted in a
proved difficult to extract product 20 from water. When the 25:75 mixture of E/Z-1c. A crystallographic analysis of crystals
reaction was performed in acetone this allowed for a simple of Z-25 confirmed the nature and geometry of the obtained
filtration of the sodium chloride and bromide salts formed and product (see Supporting Information File 2). Addition of carb-
resulted in very satisfactory yields (91–97%) after column chro- anions or alcoholates was also attempted but the starting bro-
matography. An Arbuzov phosphonylation was performed on mide E-1d degraded under these conditions.
the crude lactone product 15 (containing 10% of ester Z-13) and
proved successful with phosphonolactone 21 being isolated in Finally, this expeditious synthesis of tetrasubstituted fluoro-
77% yield after column chromatography (Scheme 7). To access alkenes by sequential ring-opening and nucleophilic substitu-
tetrasubstituted alkenes, reduction of 21 to generate diol 22 was tion reactions was applied to test the robustness of a selective
explored with lithium borohydride in Et2O. However, the reac- preparation of precursors of ACN (VII) bearing different func-
tion was slow at 20 °C and did not progress beyond 50% tional groups (Scheme 9). A particular focus was applied to the
conversion even after the addition of excess LiBH4. After preparation of the phosphonate 29, a precursor of VII that is not
purification by flash chromatography starting lactone 21 was accessible from diol VIII.
obtained in 49% yield and the desired diol 22 in 47% yield.
When the reaction was carried out in refluxing THF, a com- First, starting from pure alkene E-9, the introduction of a
plete conversion was achieved but also with impurities. This protected alcohol as a mimic of the naturally occurring
Scheme 7: Functionalization of bromomethyllactone 15 and its analogues.
1943

Beilstein J. Org. Chem. 2020, 16, 1936–1946.
Scheme 8: Functionalization by substitution reaction of the bromide E-1d vs ring-opening reaction of the oxetane 1.
Scheme 9: Preparation of tetrasubstituted fluoroalkenes.
3’-hydroxy group was achieved by allylic bromine displace- 27 with LiBr (5 equiv), to give the phosphonate 29 in 76%
ment with AcOK to efficiently afford alkene 26. The phos- overall yield. Finally, azide 28 was obtained in 89% yield in
phonate was introduced in three steps through the formation of two steps from the non-isolated intermediate mesylate 27. After
intermediate mesylate 27. This mesylate was progressed with- deacetylation, 28 was readily converted to E-24 (see Scheme 8).
out purification, albeit contaminated (10%) with the corre- These transformations of alkene E-9 illustrated how the geome-
sponding chloride (not shown). An Arbuzov reaction was per- try can be controlled for the preparation of tetrasubstituted
formed directly on the allylic bromide obtained by treatment of fluoroalkenes. The synthesis of nucleotide mimics from either
1944

Beilstein J. Org. Chem. 2020, 16, 1936–1946.
2. Purser, S.; Moore, P. R.; Swallow, S.; Gouverneur, V. Chem. Soc. Rev.
2008, 37, 320–330. doi:10.1039/b610213c
phosphonate 29 or azide 28 is underway and will be reported in
due course.
3. Gillis, E. P.; Eastman, K. J.; Hill, M. D.; Donnelly, D. J.; Meanwell, N. A.
J. Med. Chem. 2015, 58, 8315–8359.
Conclusion
doi:10.1021/acs.jmedchem.5b00258
The selective synthesis of tetrasubstituted E- or Z-fluoroalkenes
was achieved by ring-opening reactions of fluoroalkylidene-
oxetanes, with the presence of the fluorine atom governing
regioisomeric attack of the bromide ion. Functionalization of
the resultant bromoalcohols with nucleophiles led, in two steps
from oxetanes, to a series of highly functionalized tetra-
substituted fluoroalkenes with excellent geometric control. This
method offers ready access to novel fluoroalkenes as potential
precursors of important drug mimics.
4. O’Hagan, D. J. Fluorine Chem. 2010, 131, 1071–1081.
doi:10.1016/j.jfluchem.2010.03.003
5. Len, C.; Mackenzie, G. Tetrahedron 2006, 62, 9085–9107.
doi:10.1016/j.tet.2006.07.050
6. Bassetto, M.; Slusarczyk, M. Pharm. Pat. Anal. 2018, 7, 277–299.
doi:10.4155/ppa-2018-0028
7. Xie, M.-S.; Niu, H.-Y.; Qu, G.-R.; Guo, H.-M. Tetrahedron Lett. 2014,
55, 7156–7166. doi:10.1016/j.tetlet.2014.11.060
8. De Clercq, E. Med. Res. Rev. 2013, 33, 1278–1303.
doi:10.1002/med.21283
9. Schramm, V. L. Chem. Rev. 2018, 118, 11194–11258.
doi:10.1021/acs.chemrev.8b00369
10.Kasthuri, M.; El Amri, C.; Lefort, V.; Périgaud, C.; Peyrottes, S.
New J. Chem. 2014, 38, 4736–4742. doi:10.1039/c4nj00813h
11.Topalis, D.; Pradère, U.; Roy, V.; Caillat, C.; Azzouzi, A.; Broggi, J.;
Snoeck, R.; Andrei, G.; Lin, J.; Eriksson, S.; Alexandre, J. A. C.;
El-Amri, C.; Deville-Bonne, D.; Meyer, P.; Balzarini, J.; Agrofoglio, L. A.
J. Med. Chem. 2011, 54, 222–232. doi:10.1021/jm1011462
12.Amblard, F.; Nolan, S. P.; Schinazi, R. F.; Agrofoglio, L. A. Tetrahedron
2005, 61, 537–544. doi:10.1016/j.tet.2004.11.019
13.Varada, M.; Erande, N. D.; Kumar, V. A. RSC Adv. 2015, 5,
97824–97830. doi:10.1039/c5ra15673d
Supporting Information
The experimental section describing the preparation of all
new compounds, the copies of the NMR data (1H NMR,
13C NMR, 19F NMR), HOESY and NOESY experiments
and crystallographic data for compounds 1d, 4d and 25.
The CIF files of 1d, 4d and 25.
Supporting Information File 1
14.Champagne, P. A.; Desroches, J.; Paquin, J.-F. Synthesis 2015, 47,
306–322. doi:10.1055/s-0034-1379537
Experimental section and copies of spectra.
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-16-160-S1.pdf]
15.Choi, M.-H.; Lee, C.-K.; Jeong, L. S.; Chun, M. W.; Kim, H.-D.
Nucleosides, Nucleotides Nucleic Acids 2001, 20, 681–684.
doi:10.1081/ncn-100002350
Supporting Information File 2
Crystallographic data (cif) for compounds E-1d, E-4d, and
Z-25.
16.Choi, M.-H.; Kim, H.-D. Arch. Pharmacal Res. 1997, 20, 501–506.
doi:10.1007/bf02973948
17.Prunier, A.; Calata, C.; Legros, J.; Maddaluno, J.; Pfund, E.;
Lequeux, T. J. Org. Chem. 2013, 78, 8083–8097.
doi:10.1021/jo401356j
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-16-160-S2.zip]
18.Drouin, M.; Hamel, J.-D.; Paquin, J.-F. Synthesis 2018, 50, 881–955.
doi:10.1055/s-0036-1591867
19.Landelle, G.; Bergeron, M.; Turcotte-Savard, M.-O.; Paquin, J.-F.
Chem. Soc. Rev. 2011, 40, 2867–2908. doi:10.1039/c0cs00201a
20.Yanai, H.; Taguchi, T. Eur. J. Org. Chem. 2011, 5939–5954.
doi:10.1002/ejoc.201100495
Funding
This work was supported by the excellence laboratory LabEx
SYNORG (ANR-11-LABX-0029), ANR funding DALATAR
(ANR-19-CE18-0008-03) the Conseil Régional de Normandie
and the European FEDER funding.
21.Pfund, E.; Lequeux, T.; Gueyrard, D. Synthesis 2015, 47, 1534–1546.
doi:10.1055/s-0034-1380548
22.Laporte, R.; Prunier, A.; Pfund, E.; Roy, V.; Agrofoglio, L. A.;
Lequeux, T. Eur. J. Org. Chem. 2015, 3121–3128.
doi:10.1002/ejoc.201500172
ORCID® iDs
23.Yadav, J. S.; Singh, V. K.; Srihari, P. Org. Lett. 2014, 16, 836–839.
doi:10.1021/ol403604u
Clément Q. Fontenelle - https://orcid.org/0000-0002-1630-3407
Thibault Thierry - https://orcid.org/0000-0003-0342-2391
Romain Laporte - https://orcid.org/0000-0002-5984-6794
Emmanuel Pfund - https://orcid.org/0000-0003-1896-0109
Thierry Lequeux - https://orcid.org/0000-0003-3783-4458
24.Burkhard, J. A.; Wuitschik, G.; Rogers-Evans, M.; Müller, K.;
Carreira, E. M. Angew. Chem., Int. Ed. 2010, 49, 9052–9067.
doi:10.1002/anie.200907155
25.Wang, Z.; Chen, Z.; Sun, J. Angew. Chem., Int. Ed. 2013, 52,
6685–6688. doi:10.1002/anie.201300188
26.Burkhard, J.; Carreira, E. M. Org. Lett. 2008, 10, 3525–3526.
doi:10.1021/ol801293f
References
1. Zhou, Y.; Wang, J.; Gu, Z.; Wang, S.; Zhu, W.; Aceña, J. L.;
Soloshonok, V. A.; Izawa, K.; Liu, H. Chem. Rev. 2016, 116, 422–518.
doi:10.1021/acs.chemrev.5b00392
27.Jaśkowska, J.; Kowalski, P. J. Heterocycl. Chem. 2008, 45,
1371–1375. doi:10.1002/jhet.5570450519
1945

Beilstein J. Org. Chem. 2020, 16, 1936–1946.
28.Wolin, R.; Connolly, M.; Afonso, A.; Hey, J. A.; She, H.; Rivelli, M. A.;
Willams, S. M.; West, R. E., Jr. Bioorg. Med. Chem. Lett. 1998, 8,
2157–2162. doi:10.1016/s0960-894x(98)00379-5
29.Calata, C.; Pfund, E.; Lequeux, T. J. Org. Chem. 2009, 74, 9399–9405.
doi:10.1021/jo901540c
30.Fort, D. A.; Woltering, T. J.; Alker, A. M.; Bach, T. Heterocycles 2014,
88, 1079–1100. doi:10.3987/com-13-s(s)67
31.White, A. R.; Kozlowski, R. A.; Tsai, S.-C.; Vanderwal, C. D.
Angew. Chem., Int. Ed. 2017, 56, 10525–10529.
doi:10.1002/anie.201704119
32.Fort, D. A.; Woltering, T. J.; Nettekoven, M.; Knust, H.; Bach, T.
Chem. Commun. 2013, 49, 2989–2991. doi:10.1039/c3cc40757h
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0). Please note
that the reuse, redistribution and reproduction in particular
requires that the authors and source are credited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(https://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.16.160
1946