Di- tert-butyl Phosphonate Route to the Antiviral Drug Tenofovir

Article abstract of DOI:10.1021/acs.oprd.0c00473

Di-tert-butyl oxymethyl phosphonates were investigated regarding their suitability for preparing the active pharmaceutical ingredient tenofovir (PMPA). First, an efficient and simple access to the crystalline di-tert-butyl(hydroxymethyl)phosphonate was developed. O-Mesylation gave high yields of the active phosphonomethylation reagent. For the synthesis of tenofovir, a two-step sequence was developed using Mg(OtBu)2 as the base for the alkylation of (R)-9-(2-hydroxypropyl)adenine. Subsequent deprotection could be achieved with aqueous acids. (Di-tert-butoxyphosphoryl)methyl methanesulfonate showed to be the most efficient electrophile tested, affording PMPA in 72% yield on a 5 g scale. The developed protocol could also be applied for the preparation of the hepatitis B drug adefovir (64% yield/1 g scale).

Full text of DOI:10.1021/acs.oprd.0c00473

This article is made available via the ACS COVID-19 subset for unrestricted RESEARCH re-use
and analyses in any form or by any means with acknowledgement of the original source.
These permissions are granted for the duration of the World Health Organization (WHO)
declaration of COVID-19 as a global pandemic.
pubs.acs.org/OPRD
Article
Di-tert-butyl Phosphonate Route to the Antiviral Drug Tenofovir
Jule-Philipp Dietz, Dorota Ferenc, Timothy F. Jamison, B. Frank Gupton, and Till Opatz*
Cite This: Org. Process Res. Dev. 2021, 25, 789−798
Read Online
ACCESS
Metrics & More
Article Recommendations
*
sı Supporting Information
ABSTRACT: Di-tert-butyl oxymethyl phosphonates were investigated regarding their suitability for preparing the active
pharmaceutical ingredient tenofovir (PMPA). First, an efficient and simple access to the crystalline di-tert-butyl(hydroxymethyl)-
phosphonate was developed. O-Mesylation gave high yields of the active phosphonomethylation reagent. For the synthesis of
t
tenofovir, a two-step sequence was developed using Mg(O Bu) as the base for the alkylation of (R)-9-(2-hydroxypropyl)adenine.
2
Subsequent deprotection could be achieved with aqueous acids. (Di-tert-butoxyphosphoryl)methyl methanesulfonate showed to be
the most efficient electrophile tested, affording PMPA in 72% yield on a 5 g scale. The developed protocol could also be applied for
the preparation of the hepatitis B drug adefovir (64% yield/1 g scale).
KEYWORDS: tenofovir, phosphites, oxymethyl phosphonates, adefovir, antivirals
1
. INTRODUCTION
scoping the two steps proved to be beneficial to circumvent the
inconvenient workup and the continuous extraction of the
water-soluble phosphonate ester 6. Furthermore, the costs of
the deprotection step could be reduced by replacing the
The nucleotide reverse transcriptase inhibitor tenofovir (1,
PMPA), which was described by Balzarini in 1993, currently
belongs to the most frequently applied human immunodefi-
ciency virus (HIV) medications. To enable oral application
and to increase bioavailability, a prodrug unit is required,
which led to the development of tenofovir disoproxil fumarate
1
8
hitherto used expensive TMSBr by TMSCl/NaBr. However,
the workup was still elaborate, as several filtration and
extraction steps were required, while simultaneously avoiding
any moisture. Despite a reported significant loss of the product
(
2, TDF), which was approved by the FDA for HIV treatment
2
in the magnesium salt cake (up to 15%), PMPA·H O could be
in 2001 and later in 2008 for hepatitis B virus therapy. In
010, tenofovir alafenamide fumarate (3, TAF) (Figure 1) was
2
isolated in 59% yield.
In 2016, Riley et al. reported that Mg(O Bu) could be
replaced by in situ generation from MeMgCl and BuOH. The
2
t
launched, which showed fewer side effects and better
tolerability than TDF.
In the course of the 2020 COVID-19 pandemic, testing
tenofovir as potential medication for SARS-CoV2 also moved
in the focus of attention. It could be demonstrated that the
triphosphate of tenofovir inhibits the RNA-dependent RNA
2
3
t
group developed an improved method to isolate the
phosphonate diester 6 after continuous extraction in 85%
yield. Subsequent deprotection with HBr/acetic acid afforded
9
PMPA·H O in 67% yield (57% over two steps) (Scheme 1).
2
4
polymerase of this virus in vitro. These results were supported
Recently, Derstine et al. published a new approach for making
PMPA on a multigram scale. They found conditions to alkylate
HPA with the free phosphonic acid of DESMP by using
by a study, which investigated the incidence and severity of
COVID-19 from 77,590 HIV-positive persons receiving anti-
retroviral therapy. There was a lower risk of COVID-19, less
related hospitalization, and even no mortality for people who
were treated with Truvada [TDF + FTC (emtricitabine)] than
for people who had taken other antiviral drugs. There is also
an ongoing clinical study in Spain with 4000 medical workers
t
NaO Bu as a base in dimethylformamide (DMF). This very
10
efficient process gave PMPA in 70% yield.
It is conspicuous that all reports investigating the challenging
synthesis of PMPA focused only on DESMP or the derived
free acid 7 as alkylating agents with HPA. One big selling point
is of course the fact that DESMP can be made from cheap
commodity chemicals and has an industrial well-established
process. Nevertheless, the overall yield starting from diethyl
phosphite (8) has been reported to be only 60−70% with
5
6
investigating Truvada as potential prophylaxis for COVID-19.
Thus, the demand for tenofovir could further rise in the future.
Because of a lack of a diverse set of industrial syntheses of
tenofovir, raw material dependency can lead to unsteady prices
and drug shortages. Therefore, diversification of the synthetic
portfolio is an attractive goal.
11
varying degrees of purity (Scheme 2).
The state-of-the-art synthesis of PMPA has been reported by
Ripin et al. in 2010 (Scheme 1). (R)-9-(2-Hydroxypropyl)-
adenine (HPA, 4) is alkylated with diethyl(p-
toluenesulfonyloxymethane)phosphonate (DESMP, 5), fol-
7
Received: October 22, 2020
Published: February 11, 2021
lowed by a one-pot deprotection of the phosphonic acid
t
ester. Ripin et al. reported that using Mg(O Bu) as a base
2
showed the best conversion (>90%) for alkylation. Tele-
©
2021 American Chemical Society
https://dx.doi.org/10.1021/acs.oprd.0c00473
Org. Process Res. Dev. 2021, 25, 789−798
7
89

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Figure 1. Structures of tenofovir (PMPA, 1), TDF (2), and TAF (3).
Scheme 1. Overview to Hitherto Known Scalable PMPA Syntheses
Scheme 2. Industrial Synthesis of DESMP (5)
filtrations and extractions is necessary. Silicon-containing
12
byproducts have also been reported. Other conditions (aq.
13
9
HBr, dry HCl gas, AlCl3, or HOAc/HBr ) led to lower
yields. There is no variation in the alkyl group of oyxmethyl
phosphonates reported in the literature other than the use of
diisopropyl ester, which requires similar conditions for the
1
4,15
deprotection as the diethyl ester 6.
Tert-butyl phosphonates
16,17
are also known for being
The biggest drawback of using DESMP in the synthesis of
PMPA remains the deprotection of the diethyl ester. Besides
deprotected under aqueous acidic conditions. In this report,
the synthesis of di-tert-butyl oxymethyl phosphonates and their
suitability for preparing PMPA was investigated. An industri-
ally feasible process was the aim of this work (Scheme 3).
6
expensive (TMSBr ) or excessive amounts of reagents
7
(
TMSCl/NaBr ), an elaborate workup including several
Scheme 3. Proposed Synthesis for PMPA Using di-tert-butyl Oxymethyl Phosphonates
7
90
https://dx.doi.org/10.1021/acs.oprd.0c00473
Org. Process Res. Dev. 2021, 25, 789−798

Organic Process Research & Development
. RESULTS AND DISCUSSION
.1. Synthesis of di-tert-butyl Phosphite. Di-tert-butyl
pubs.acs.org/OPRD
Article
t
2
Table 1. Screening Equivalents of KO Bu for the Synthesis
of di-tert-butyl Phosphite
a
2
phosphite (9) is commercially available but expensive. It can
be synthesized by adding PCl to a cooled solution of tert-
3
butanol in the presence of a base (triethylamine, pyridine, or
dimethylaniline) in a nonpolar solvent (ligroin, petroleum
ether, and diethyl ether). The reported yields vary widely (40−
t
b
c
#
KO Bu (equiv)
IY [%]
purity [%]
1
8−23
7
7%).
Chemoselectivity (triester 10 vs diester 9, Figure 2)
1
2
3
4
5
3.0
2.9
2.7
2.5
2.3
98
85
84
76
69
75
91
95
97
99
24
also proved to be an issue.
a
b
All reactions were performed on a 3 g scale. Crude isolated yield.
c
31
Determined by P NMR-spectroscopy.
Using the water-soluble THF as the solvent required large
amounts of sodium sulfate for drying. Therefore, less polar
alternative solvents were investigated. CPME (cyclopentyl
methyl ether), 2-Me-THF, and even MTBE showed similar
results regarding yield and purity (Table 2).
Figure 2. Structures of di-tert-butyl phosphite (9), tri-tert-butyl
phosphite (10), and tri-tert-butyl phosphate (11).
Mark and Van Wazer, who focused their investigations on
the selective synthesis of tri-tert-butyl phosphite (10),
Table 2. Solvent Screening for the Synthesis of di-tert-butyl
Phosphite
enlightened more details about the reaction of PCl and tert-
a
3
butanol. They proved that almost only triester 10 forms in the
reaction. Nevertheless, 10 decomposes quickly at 50 °C under
reduced pressure yielding the diester 9. Furthermore, they
reported that triester 10 was prone to rapid aerial oxidation to
tri-tert-butyl phosphate (11, Figure 2). In terms of the selective
synthesis of the diester, 11 represents an undesired byproduct,
which can only be avoided by strict exclusion of oxygen during
preparation and workup. The reported varying yields could be
explained by the fact that diester 9 has been reported to be
very sensitive to incautious heating and some product is lost
t
b
c
#
solvent
THF
2-Me-THF
CPME
KO Bu (equiv)
IY [%]
purity [%]
1
2
3
4
2.5
2.5
2.5
2.5
76
75
77
97
99
99
99
MTBE
78
a
b
All reactions were performed on 3 g scale. Crude isolated yield.
c
31
Determined by P NMR spectroscopy.
19,20
during distillation.
In order to make tert-butyl phosphite
By focusing on MTBE and 2-Me-THF as solvents,
t
interesting for an industrial application, the development of a
substitution of KO Bu by the cheaper sodium salt further
more efficient synthesis was attempted.
It appeared worth trying the reaction of KO Bu with PCl
instead of tert-butanol and an organic base. As a benefit, KO Bu
increased the yield of 9 (Table 3). When the reaction in
MTBE was scaled up, an increasing byproduct formation was
observed, whereas in Me-THF, the purity remained constant.
The obtained crude di-tert-butyl phosphite was used for
hydroxymethylation without prior distillation to avoid an
additional loss of products.
t
3
t
would combine both the reagent and base, while only
potassium chloride would be accumulated as a coproduct. In
t
initial experiments, PCl was added to a KO Bu/tetrahydrofur-
3
an (THF) solution while cooling. Following the reaction by
2.2. Hydroxymethylation of di-tert-butyl Phosphite.
1
31
H/ P NMR spectroscopy also revealed the formation of
A procedure for hydroxymethylation of di-tert-butyl phosphite
2
5,26
triester, diester, and phosphate. However, when the order of
addition was reversed, predominant diester formation was
observed and none of the potential impurities 10 and 11 could
has already been reported in the patent literature.
According to this procedure, 9 was stirred in the presence of
aq. formaldehyde solution, triethylamine, and water at r.t. to
31
31
be detected using P NMR spectroscopy. Furthermore, adding
give crude 12 in 99% yield. A P NMR spectrum showed
t
solid KO Bu instead of a THF solution, which was slightly
about 10% of impurities (Scheme 4).
more convenient regarding handling, gave similar results. For
the workup, the reaction mixture was quenched with saturated
For the workup, it was necessary to coevaporate the reaction
mixture several times with MeOH and DCM in order to
remove water and triethylamine. Optimization attempts by
replacing triethylamine by K CO or omitting the additional
NaHCO solution to maintain a basic pH in order to prevent
3
acid-catalyzed cleavage of the tert-butyl groups. It is worth
mentioning that despite a second aqueous washing step, all the
product remained in the organic phase and no further
2
3
water led to increased byproduct formation. In search for a
simpler method, the combination of paraformaldehyde with
K CO in acetonitrile showed promising results. After
t
extraction was necessary. Using 3.0 equiv of KO Bu, 98% of
2
3
3
1
crude 9 ( P NMR purity: 75%) was obtained after workup
Table 1, entry 1). Further investigations revealed that
performing some optimization studies (see the Supporting
Information for more details), nearly pure 12 could be isolated
in 99% yield after workup, which only consisted of a single
filtration and subsequent solvent removal in vacuo (Scheme 5).
It should be mentioned that only commercial or freshly
distilled di-tert-butyl phosphite was used up to this point.
When using the crude di-tert-butyl phosphite obtained through
the newly developed protocol mentioned above, the purity was
(
t
subsequent reduction of the equivalents of KO Bu improved
the purity of the crude product (Table 1, entries 2−5). When
t
the reaction was performed with 2.5 equiv of KO Bu, 76% of
1
31
crude 9 could be obtained showing a high purity in H and P
NMR spectroscopies (see the Supporting Information for
details).
7
91
https://dx.doi.org/10.1021/acs.oprd.0c00473
Org. Process Res. Dev. 2021, 25, 789−798

Organic Process Research & Development
pubs.acs.org/OPRD
Article
t
Table 3. Scale Up of the di-tert-butyl Phosphite Synthesis Using NaO Bu
t
a
b
#
scale [grams of PCl3]
solvent
MTBE
MTBE
2-Me-THF
2-Me-THF
NaO Bu (equiv)
IY [%]
purity [%]
1
2
3
4
3
6
3
2.5
2.5
2.5
2.5
80
82
86
85
97
90
95
95
9
a
Crude isolated yield. Determined by 31P NMR spectroscopy.
b
Scheme 4. Hydroxymethylation of di-tert-butyl Phosphite
Using aq. Formaldehyde Solution
was efficient to fill the gas room of the flask and the balloon
with argon. Paraformaldehyde precipitation could not be
observed anymore, and consequently, the equivalents of
paraformaldehyde could be further decreased to 1.0 equiv,
while temperature and time were also adjusted. Crude 9
derived from the esterification in MTBE could be hydrox-
ymethylated in 66% yield (Table 4, entry 2), while crude 9
derived from the esterification in Me-THF furnished 12 in 72%
(Table 4, entry 3). Upscaling to an 11 g reaction afforded 12 in
71% yield. There are still opportunities for optimization of this
two-step sequence. With a more accurate method for
determining the purity of crude di-tert-butyl phosphite (e.g.,
gas chromatography), the equivalents of paraformaldehyde
could be adjusted more precisely to produce less byproducts.
Scheme 5. Hydroxymethylation of di-tert-butyl Phosphite
Using Paraformaldehyde
2.3. Tosylation and Mesylation of 12. Sulfonylation of
the hydroxyl group of 12 has only been described for the
expensive triflate. Tosylation and mesylation of 12 should be
investigated as more economic alternatives. Under optimized
conditions, (di-tert-butoxyphosphoryl)methyl 4-methylbenze-
nesulfonate (13) could be obtained in 87% yield after
recrystallization from EtOH (Scheme 6). It should be noted
determined first by 31P NMR spectroscopy in order to adjust
the amount of paraformaldehyde. This was crucial, as excess
paraformaldehyde caused additional byproduct formation.
During further investigations, it turned out that 12 can be
recrystallized from MeCN. When the filtered and concentrated
reaction mixture was stored in a freezer, crystalline 12 could be
25
isolated in 71% (based on PCl ) (Table 4, entry 1).
Scheme 6. Tosylation of di-tert-
3
butyl(hydroxymethyl)phosphonate
Table 4. Telescoped Synthesis of di-tert-
butyl(hydroxymethyl)phosphonate from PCl3
a
b
#
scale [grams of 9]
(CH O)n (equiv) T [°C] t [h] IY [%]
2
that 13 is prone to rapid dealkylation in the presence of acids.
When the reaction was performed without DMAP, traces of
tosyl chloride present in the crude product were sufficient to
slowly release HCl during recrystallization in EtOH, which led
to dealkylation.
The mesylation of 12 proceeded more efficiently than the
tosylation under similar conditions. Furthermore, no recrystal-
lization was necessary, as the excess of mesyl chloride
completely hydrolyzed during the aqueous workup and no
byproducts formed. Crystalline (di-tert-butoxyphosphoryl)-
methyl methanesulfonate (14) could be obtained in 97%
yield even on a multigram scale (Scheme 7).
1
6.6 (95)
7.2 (90)
3.8 (95)
11.3 (95)
1.1
1.0
1.0
80
70
70
70
5
20
20
20
71
66
72
71
c
c
c
2
3
4
1.0
31
a
b
Purity [%] determined by P NMR spectroscopy. After
recrystallization and based on PCl . Flask gas room filled with argon.
c
3
Because of the volatility of the releasing formaldehyde, the
reactions were usually performed in a closed flask equipped
with a septum and a nitrogen-filled balloon for pressure
compensation. However, there were always small amounts of
paraformaldehyde precipitating on the bottom of the septum.
Therefore, a slight excess of paraformaldehyde (0.1 equiv) had
to be used. Depending on the scale and reactor size, the
precipitating paraformaldehyde could lead to variable con-
version and made unwelcome additional purification necessary.
To circumvent this issue at least under laboratory conditions, it
2.4. Synthesis of PMPA Using Di-tert-butyl Oxy-
methyl Phosphonates 13 and 14. For screening the
alkylation of HPA, the focus was laid on using mesylate 14
because of its more efficient preparation. Performing the
t
reaction in DMF and using other bases than Mg(O Bu) such
2
t
t
as KO Bu, NaO Bu, or NaH led to product mixtures containing
7
92
https://dx.doi.org/10.1021/acs.oprd.0c00473
Org. Process Res. Dev. 2021, 25, 789−798

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Scheme 7. Mesylation of di-tert-
butyl(hydroxymethyl)phosphonate
with NaOH solution to precipitate PMPA·H O, which was
2
10
dried afterward, as described in the literature. On a 1 g scale,
the conversion of HPA was slightly higher when using 1.7
equiv of 14 instead 1.5 equiv, while the isolated yields were
similar (64 and 65%, respectively, Table 5, entries 1 and 2).
Despite the high conversions of HPA, large amounts of PMPA·
H O did not precipitate and remained in the mother liquor.
2
Several attempts were undertaken to isolate more PMPA
from the mother liquor. When the mother liquor was
concentrated and cooled again, mostly inorganic salts
precipitated along with traces of PMPA. The same was
observed when small amounts of water were added to the
residue of the lyophilized mother liquor. The addition of
both N-and O-alkylated HPA, bis-alkylated HPA, and both
isomers of 9-propenyladenine, which probably had formed
because of transesterification and subsequent elimination. The
t
best results were achieved with Mg(O Bu) as the base. In the
2
i
cosolvents like MeOH, EtOH, PrOH, or MeCN to the mother
first screening, different aprotic polar solvents [DMF, dimethyl
sulfoxide (DMSO), dimethyl acetamide (DMA), and N-
methylpyrrolidone (NMP)] were investigated, while an excess
of mesylate 14 (2.5 equiv) was used to accomplish a complete
conversion of HPA.
The best results were observed when O-alkylation was
performed in NMP or DMA at 90 °C. Mainly, monoester 16
formation occurred along with only small amounts of
byproducts. Summing up the monoester 16 and diester 15,
liquor also led only to the precipitation of PMPA/salt mixtures
from which PMPA could not be easily separated. Derstine et al.
have carefully observed that NaCl, which forms during pH
10
adjustment, has a large effect on keeping PMPA in solution.
Therefore, in a further experiment, aqueous H SO was used
2
4
for dealkylation and the pH was adjusted with conc. aqueous
NH (25 wt %) in order to see how (NH ) SO formation
3
4 2
4
affects the solubility of PMPA. Only slightly more PMPA
69%) could be isolated in this way (Table 5, entry 3).
(
9
5 and 96% (Figure 3) PMPA formation could be detected,
Nevertheless, performing the reaction on a 5 g scale, 72% of
PMPA could be obtained (Table 5, entry 4).
Using the tosylate 13 for the alkylation of HPA, similar
conversions could be observed in DMA, DMSO, and DMF
respectively (see the Supporting Information for further
details).
For further optimizations, DMA was chosen as the solvent.
t
Regarding the amount of Mg(O Bu) , it turned out that 3.0
2
(
see Supporting Information for more details). When the
equiv was crucial to achieve high conversions (>90%).
Between 2.0 and 1.5 equiv of mesylate 14, the conversion
varied between 90 and 98%, and only decreased noticeably
below 1.5 equiv (see Supporting Information for further
details).
To cleave off the tert-butyl groups in order to generate
PMPA, aqueous hydrochloric acid was added to the crude
reaction mixture. Dealkylation was complete after stirring for
reaction was performed in DMF on a 1 g scale, a 96%
conversion of HPA was detected after 23 h. After conducting
dealkylation and workup in the same way as described for the
mesylate 14, PMPA could be isolated in 68% yield (Scheme
8).
All in all, there was no significant difference in the isolated
yield when using mesylate 14 or tosylate 13 for preparing
PMPA.
2
4 h at r.t. When the reaction was performed on a larger scale
(
1 g), DMA was first removed in vacuo and aqueous acid (3 N
2.5. Further Application. Both newly reported di-tert-
butyl phosphonates 13 and 14 could also be easily converted
quantitatively to the corresponding free phosphonic acids by
HCl) was added to the residue. At 60 °C, dealkylation was
already complete after 2 h. The pH was adjusted to 2.8−3.0
Figure 3. High-performance liquid chromatography (HPLC) chromatogram of the alkylation of HPA with mesylate 14 after 21 h in DMA at 90 °C
λ = 254 nm).
(
7
93
https://dx.doi.org/10.1021/acs.oprd.0c00473
Org. Process Res. Dev. 2021, 25, 789−798

Organic Process Research & Development
pubs.acs.org/OPRD
Article
Table 5. Two-Step Sequence for Making PMPA from HPA and Mesylate 14
a
#
scale [grams of 4]
14 (equiv)
t [h]
Conversion [%]
IY [%]
b
1
2
3
4
1.0
1.0
1.0
1.7
1.5
1.5
1.5
23
21
24
22
96
92
91
91
64
65
69
72
b
c
c
5.0
a
b
c
Determined by HPLC (λ = 254 nm). Dealkylation/pH adjustment performed with 3 N HCl/NaOH (40 wt %). Dealkylation/pH adjustment
performed with 3 N H SO /NH (25 wt %).
2
4
3
Scheme 8. Two-Step Sequence for Making PMPA from HPA and Tosylate 13
just heating them for a few minutes even under solvent-free
conditions (Scheme 9). In the case of tosylate 13, the method
affords an alternative to the recently reported synthesis by
Derstine et al., where the diethyl ester 5 was dealkylated with
3. CONCLUSIONS
A new practical two-step sequence for the synthesis of
tenofovir was developed using the hitherto unknown (di-tert-
butoxyphosphoryl)methyl 4-methylbenzenesulfonate (13) and
1
0
TMSBr.
(
di-tert-butoxyphosphoryl)methyl methanesulfonate (14).
These crystalline key intermediates could be synthesized in
gram amounts using straightforward chemistry and avoiding
any chromatography step. The newly developed tert-butyl
phosphite synthesis allowed for a simple and multigram
preparation of a sufficiently pure starting material. The crude
di-tert-butyl phosphite was hydroxymethylated through an
optimized protocol, whereby crystalline di-tert-butyl-
Scheme 9. Dealkylation of 13 and 14 to the Free
Phosphonic Acid by Heating in the Neat Form
(
hydroxymethyl)phosphonate could directly be obtained
from the filtered reaction mixture.
While the manufacturing costs of the tert-butyl oxymethyl
phosphonates are probably higher than those of the commonly
used ethyl derivative, higher expenses can probably be
compensated by much smoother and more economic
conditions for deprotection and by a simpler overall process.
Furthermore, the use of mesylate instead of tosylate as the
leaving group in the alkylation of HPA is more atom-
economical. The protocol developed herein allows for a
quick and efficient access to compounds showing an oxymethyl
phosphonate moiety, which can also be beneficial for
prospective research.
In general, other nucleotide-like APIs having an oxy-methyl
phosphonate moiety are industrially prepared similar to
2
7,28
PMPA.
The hepatitis B inhibitor adefovir 22, which was
29
described first by A. Holy and I. Rosenberg in 1987, only
differs in one methyl group from tenofovir. Applying the newly
developed protocol to 9-(2-hydroxyethyl)adenine (HEA, 23),
8
7% conversion could be detected after 22 h at 90 °C. After
dealkylation and workup, adefovir could be isolated in 64%
yield (Scheme 10). It should be noted that when 3 N H SO /
4
. EXPERIMENTAL SECTION
2
4
NH was used for dealkylation/pH adjustment, the formation
All employed chemicals were commercially available and used
3
of an insoluble adefovir salt was observed.
without prior purification. Anhydrous THF, 2-Me-THF,
Scheme 10. Synthesis of Adefovir 22 Using HEA and Mesylate 14
7
94
https://dx.doi.org/10.1021/acs.oprd.0c00473
Org. Process Res. Dev. 2021, 25, 789−798

Organic Process Research & Development
pubs.acs.org/OPRD
Article
CPME, and MTBE were freshly distilled over potassium
shaken vigorously. The aqueous phase was drained and the
organic phase was washed again with a fresh portion of sat.
(
DCM over CaH ) under a nitrogen atmosphere. Anhydrous
2
DMF, DMA, DMSO, and NMP were purchased from Acros
NaHCO -solution (200 mL). The organic phase was separated
3
(
AcroSeal). Oven-dried glassware was dried in an oven at 150
and dried over NaSO , and all volatiles were removed in vacuo
4
°C overnight, closed with a plug and a septum while still hot,
at 30 °C. Crude 9 was obtained as a colorless, slightly turbid
liquid (11.26 g, 57.98 mmol, 85%), which was used for the
next step without further purification. M (C H O P) = 194.21
cooled to room temperature, and then purged with nitrogen.
NMR spectra were recorded on a Bruker AVANCE-III HD
8
19
3
1
13
31
instrument ( H NMR: 300 MHz, C NMR: 75 MHz,
P
g/mol. Boiling range: 27−31 °C (0.3 mbar) (lit. 72−78 °C
1
9
NMR: 121 MHz) or a Bruker AVANCE-III HD instrument
(
13−16 mbar) ) R (SiO ): 0.46 (EtOAc), stained with
f
2
1
13
31
(
H NMR: 400 MHz, C NMR: 101 MHz, P NMR: 162
−1
1
KMnO . IR (ATR) ν: 2980, 1371, 1263, 1173, 958 cm . H
NMR, COSY (300 MHz, CDCl ): δ 6.90 (d, J
H, −P−H), 1.47 (s, 18H, −C(CH ) ) ppm. C NMR,
HMBC, HSQC (75 MHz, CDCl ): δ = 82.9 (d, J
4
MHz) with a 5 mm BBFO probe. The chemical shifts δ were
1
= 681 Hz,
3
P−H
13
1
expressed in ppm downfield from tetramethylsilane ( H NMR,
1
3
3
1
3
C NMR). Deuterated solvents (CDCl , DMSO-d ) served as
2
3
6
= 7.4
3
C−P
the internal reference. The reported signal splittings were
3
Hz, −O−C(CH ) −), 30.5 (d, J
= 4.6 Hz, −CH ) ppm.
3
3
C−P
3
abbreviated as follows: s = broad singlet, s = singlet, d =
31
b
P NMR (121 MHz, CDCl ): δ 3.18 ppm. GC−MS m/z: 83.1
3
doublet, and t = triplet. Coupling constants J are reported in
(
100%). The spectrometric data are consistent with literature
3
1
Hz. For P NMR purity analyses, the analyte (15 mg) was
dissolved in the deuterated solvent indicated (0.7 mL). ESI-
MS spectra were recorded on a 1260-series Infinity II HPLC
system (Agilent-Technologies) with a binary pump and
integrated diode array detector coupled to a liquid
chromatography/mass selective detector (LC/MSD) Infinity-
lab LC/MSD (G6125B LC/MSD) mass spectrometer. For
high-resolution (HR) mass spectra, an Agilent 6545 Q-TOF
spectrometer and a suitable external calibrant was used.
Analytical HPLC was carried out with an Agilent 1260 Infinity
system equipped with a binary pump, a diode array detector,
and LC/MSD Infinitylab LC/MSD (G6125B LC/MSD) mass
spectrometer. An ACE C18 pentafluorophenyl column (3 μm,
22,24
values.
4.2. Di-tert-butyl(hydroxymethyl)phosphonate, 12.
Variant 1: According to a method reported by Grimmond et
25
al., a round-bottomed flask was charged with di-tert-butyl
phosphite (commercially available, 96%, 3.24 g, 16.0 mmol, 1.0
equiv), NEt (2.6 mL, 19 mmol, 1.2 equiv), and H O (1 mL).
3
2
Aqueous formaldehyde solution (37%, 1.20 mL, 16.0 mmol 1.0
equiv) was added afterward, and the solution was stirred for 24
3
1
h at r.t. (reaction control by GC or P NMR spectroscopy).
MeOH (5 mL) was added, and all volatiles were removed in
vacuo at 40 °C. This procedure was repeated twice and then
performed again with DCM (3 × 5 mL), furnishing the crude
product as a slight yellowish waxy solid (3.51 g, 14.5 mmol,
4
.6 mm × 150 mm, 40 °C) with gradient elution using
3
1
acetonitrile/water (+0.1% formic acid) or phosphate buffer (20
mM, pH = 2.5)/MeOH as the solvent with a flow rate of 1.0
mL/min was used. Gas chromatography was performed on an
Agilent 8890 gas chromatograph equipped with a 5977 Gas
chromatography−mass spectrometry detector. An Agilent
Technologies HP 5MS UI column (30 m × 0.25 mm × 0.25
μm) as a stationary phase with helium as a carrier gas and a
flow rate of 1.2 mL/min was used. The following parameters
were used: inlet temperature 250 °C, transfer-line temperature
91%, yield corrected based on P NMR). Variant 2: A round-
bottomed flask was charged with crude di-tert-butyl phosphite
3
1
(95%, purity estimated by P NMR, 11.25 g, 55.0, mmol, 1.0
equiv), paraformaldehyde (97%, 1.70 g, 1.0 equiv), K CO
2
3
(anhydrous, ground, stored in a desiccator, 1.60 g, 11.59 mmol,
0.2 equiv), and MeCN (HPLC grade, 170 mL). The gas-filled
compartment of the flask was purged with argon for 15 s and
immediately closed with a septum equipped with an argon-
filled balloon. The colorless suspension was heated at 70 °C for
3
1
2
50 °C, ion-source temperature 230 °C, MS-quadrupole
temperature 150 °C, and an initial oven temperature of 40
C for 2 min with a temperature ramp of 50 °C/min to 320 °C
2
0 h (reaction control by GC or P NMR spectroscopy). The
reaction mixture was cooled to r.t., filtered, and concentrated in
vacuo at 40 °C to half of the original volume. The flask was
stored in a freezer overnight at −24 °C when 12 crystallized
out (if no crystallization took place, slight shaking or
inoculating helped). The supernatant mother liquor was
decanted, and the solid was washed twice with cold MeCN
°
over 5.6 min followed by 7.4 min hold. IR spectroscopy was
conducted on a Bruker Tensor 27 Fourier transform infrared-
spectrometer using a diamond attenuated total reflection
(
ATR) unit. Thin-layer chromatography was performed on
Merck F254 silica gel plates. Spots were visualized with UV light
λ = 254 nm) or stained with appropriate reagents. Melting
points are uncorrected and were taken by using a Kruss
(
−24 °C, 2 × 5 mL). The solid was dried in vacuo at 30 °C to
(
afford 12 (8.90 g). Concentrating the mother liquor and
̈
storing in a freezer overnight yielded a second pure crystallisate
KSP1N digital melting point apparatus. Water content
determination was conducted with a Xylem TitroLine 7500
Karl-Fischer Titrator.
(
1.97 g). In total, 10.87 g (48.47 mmol, 71% related to PCl3)
of 12 was obtained. M (C H O P) = 224.24 g/mol. mp
9
21
4
1
00.0−102.7 °C. R (SiO ): 0.32 (EtOAc), stained with
f
2
4
.1. Di-tert-butyl Phosphite, 8. An oven-dried Schlenk
KMnO . IR (ATR) ν: 3309, 2980, 1394, 1238, 1167, 1038,
4
flask was charged with dry 2-Me-THF (300 mL) under a
−
1 1
9
75 cm . H NMR, COSY (400 MHz, CDCl ): δ 3.74 (d,
3
nitrogen atmosphere and cooled in an ice bath. PCl (99%, 6.0
3
1
JP−H = 6.6 Hz, 2H, −PCH −), 2.57 (s , 1H, −OH), 1.52 (s,
2
B
mL, 67.9 mmol, 1.0 equiv) was added, and the solution was
1
3
1
8H, −C(CH ) ) ppm. C NMR, HMBC, HSQC (101 MHz,
stirred for 3 min while cooling. Under nitrogen reverse flow,
3 3
2
t
CDCl ): δ 82.9 (d, J
= 9.0 Hz, −C(CH ) −), 60.1 (d,
C−P
3
NaO Bu (98%, 16.32 g, 166.4 mmol, 2.5 equiv) was added in
3
3 3
1
JC−P = 164 Hz, −P−CH −), 30.6 (d,
J = 3.8 Hz,
C−P
small portions over 5 min while stirring vigorously. The ice
bath was removed, and the thick colorless suspension was
2
3
1
−C(CH
)
3
) ppm. P NMR (162 MHz, CDCl
): δ 16.4 ppm.
3
3
stirred for 1 h at r.t. Saturated NaHCO solution (200 mL)
GC−MS m/z: 113.1 (100%). ESI-MS m/z: 247.1 (100%, [M
3
+
was added, and the mixture was stirred for 5 min. The two-
phase mixture was transferred into a separating funnel and
+ Na] ). The spectrometric data are consistent with literature
30
values.
7
95
https://dx.doi.org/10.1021/acs.oprd.0c00473
Org. Process Res. Dev. 2021, 25, 789−798

Organic Process Research & Development
.3. (Di-tert-butoxyphosphoryl)methyl 4-Methylben-
zenesulfonate, 13. In an oven-dried Schlenk flask, di-tert-
butyl(hydroxymethyl)phosphonate (3.00 g, 13.4 mmol, 1.0
pubs.acs.org/OPRD
Article
4
HPA (4, >98%, 5.00 g, 25.9 mmol 1.0 equiv) and magnesium
tert-butoxide (93%, 14.24 g, 77.64 mmol, 3.0 equiv) under a
nitrogen atmosphere. Dry DMA (65 mL) was added, and the
suspension was stirred at 90 °C for 30 min. While purging with
nitrogen, (di-tert-butoxyphosphoryl)methyl methanesulfonate
(14, 11.74 g, 38.82 mmol, 1.5 equiv) was added portion-wise
within 1 min. The reaction mixture was stirred at 90 °C for 22
h (90% conversion of HPA detected by HPLC at 254 nm). All
volatiles were removed in vacuo at 50−60 °C, and 1.5 M
H SO (50 mL) was added to the orange-brownish residue.
equiv), NEt (2.1 mL, 14.7 mmol, 1.1 equiv), and DMAP
3
(
(
0.16 g, 1.34 mmol, 0.1 equiv) were dissolved in dry DCM
100 mL) under a nitrogen atmosphere. Tosyl chloride (2.81
g, 14.7 mmol, 1.1 equiv) was added, and the solution was
stirred under a nitrogen atmosphere at r.t. for 24 h (99%
conversion by P NMR spectroscopy). The reaction mixture
was washed with saturated NaHCO solution (70 mL), and
3
1
3
2
4
the organic phase was separated. After drying over NaSO , all
4
During heating to 60 °C, a yellowish solution formed. After 4 h
complete dealkylation to PMPA detected by HPLC at λ = 254
nm), the solution was cooled in an ice bath. Additional water
volatiles were removed in vacuo at 30 °C [CAUTION! High
temperatures can lead to decomposition (dealkylation)]. The
crude product (4.93 g) was dissolved in warm EtOH (17 mL,
max 50 °C), cooled to r.t., and stored in a freezer (−24 °C)
overnight. The supernatant mother liquor was decanted, and
the crystallized solid was washed with two portions of cold
(
(
20 mL) was added, and the pH was adjusted to pH = 2.8−3
using conc. NH solution (25 wt %, 3 mL). A colorless to slight
3
yellowish suspension formed, which was stirred while cooling
for 1 h and then stored in a refrigerator overnight. The next
day, the suspension was vacuum-filtered and washed with ice-
cold water (4 × 4 mL) and ice-cold acetone (3 × 4 mL). The
filtered solid was first dried on the air and then at 80 °C under
high vacuum for 4 h to obtain PMPA as a colorless powder
(
1
−24 °C) EtOH (2 × 3 mL). After drying in vacuo at 30 °C,
3 was obtained as a colorless solid in 87% yield (4.41 g, 11.7
mmol, 87%). For longer storage, it is recommended to store
the compound in a refrigerator or freezer. M (C H O PS) =
16
27
6
3
0
1
78.42 g/mol. mp 74.4−76.5 °C (decomposition). R (SiO ):
f
2
(
5.36 g, 18.6 mmol, 72%, HPLC-purity (254 nm): ≥99%,
q- P NMR assay: (98.37 ± 1.95)%). M (C H N O P) =
.21 (EtOAc/cyclohexane = 1:2). IR (ATR) ν: 2979, 1365,
31
−
1
1
9
14
5
4
251, 1180, 1170, 1022, 972 cm . H NMR, COSY (300
2
2
87.22 g/mol. mp 271−274 °C (decomposition) (lit. 276−
MHz, CDCl ): δ 7.82−7.77 (m, 2H, H-2), 7.37−7.33 (m, 2H,
H-3), 4.01 (d, J
CH ), 1.46 (s, 18H, −(CH ) ) ppm. C NMR, HMBC,
HSQC (101 MHz, CDCl ): δ 145.3 (C-1), 132.2 (C-4), 130.0
31
3
78 °C ). Water content: 2.3% (determined by Karl-Fischer
2
= 9.9 Hz, 2H, −P−CH ), 2.45 (s, 3H, Ar-
H−P
2
Titrator). R (C −SiO ): 0.38 (H O). IR (ATR) ν: 3383,
13
f
18
2
2
3
3 3
3
216, 3108, 2933, 1696, 1666, 1616, 1410, 1237, 1074, 933
3
−1 1
2
cm . H NMR, COSY (300 MHz, DMSO-d
1H, H-8), 8.15 (s, 1H, H-2), 7.43 (s , 2H, −NH
H−), 4.16 (dd, J = 14.3
H−), 3.95−3.85 (m, 1H,
6
): δ = 8.17 (s,
), 4.29 (dd, J
2
(
(
C-3), 128.3 (C-2, 84.3 (d, J
d, J
= 8.6 Hz, −C(CH ) ), 64.1
C−P
3
3 3
= 4.0 Hz,
2
1
B
= 174 Hz, −P−CH −), 30.4 (d, J
3
C−P
2
C−P
3
2
31
= 14.3 Hz, J = 4.0 Hz, 1H, −NCH
a
−
C(CH ) ), 21.8 (s, Ar-CH ) ppm. P NMR (121 MHz,
3
3
3
+
Hz, J = 5.6 Hz, 1H, −NCH
b
CDCl ): δ 6.7 ppm. ESI-HRMS: calcd for [M + Na] , m/z:
3
3
−
=
CH(CH )O−), 3.66−3.51 (m, 2H, −OCH P−), 1.02 (d, J
3
2
4
01.1158; found, m/z: 401.1154.
.4. (Di-tert-butoxyphosphoryl)methyl Methanesul-
fonate, 14. In an oven-dried Schlenk flask, di-tert-butyl-
hydroxymethyl)phosphonate (12.94 g, 57.71 mmol, 1.0
13
6.2 Hz, 3H, −CH ) ppm. C NMR, HMBC, HSQC (75
3
4
MHz, DMSO-d ): δ 155.5 (C-6), 151.7 (C-2), 149.7 (C-4),
6
3
1
−
(
41.9 (C-8), 118.2 (C-5), 75.3 (d, JC−P = 12.1 Hz,
(
1
CH(CH )O−), 64.5 (d, J
= 162 Hz, −OCH P−), 46.5
C−P
31
equiv) was dissolved in dry DCM (200 mL) under a nitrogen
3
2
−NCH −), 17.0 (−CH ) ppm. P NMR (121 MHz,
atmosphere. NEt (9.70 mL, 69.2 mmol, 1.2 equiv) was added,
2
3
3
DMSO-d ): δ 16.2 ppm. ESI-MS m/z: 288.1 (100%, [M +
and the solution was cooled in an ice bath. Mesyl chloride
6
+
H] ). The spectrometric data are consistent with literature
(
4.90 mL, 63.5 mmol, 1.1 equiv) was added drop-wise to the
1
0
values.
solution within 4 min, the ice bath was removed, and the
solution was stirred at r.t. for 19 h (reaction control by thin-
layer chromatography). The reaction mixture was washed
4
.5.2. Variant 2 Using Tosylate 13. An oven-dried Schlenk
flask was charged with HPA (4, >98%, 1.00 g, 5.18 mmol, 1.0
equiv) and magnesium tert-butoxide (93%, 2.85 g, 15.5 mmol,
twice with sat. NaHCO solution (2 × 150 mL), and the
3
3
.0 equiv) under a nitrogen atmosphere. Dry DMF (12 mL)
organic phase was separated. After drying over Na SO , all
2
4
was added, and the suspension was stirred at 80 °C for 25−30
min. While purging with nitrogen, (di-tert-butoxyphosphoryl)-
methyl 4-methylbenzenesulfonate (13, 2.94 g, 7.76 mmol, 1.5
equiv) was added portion-wise within 1 min. The reaction
mixture was stirred at 80 °C for 23 h (95% conversion of HPA
detected by HPLC at λ = 254 nm). All volatiles were removed
in vacuo, and 1.5 M H SO (10 mL) was added to the orange-
volatiles were removed in vacuo at 30 °C [CAUTION! High
temperatures can lead to decomposition (dealkylation)]. The
title compound 14 was obtained as an orange-brown oil (16.92
g, 55.97 mmol, 97%), which solidified after a while. For longer
storage, it is recommended to store the compound in a
refrigerator or freezer. M (C H O PS) = 302.32 g/mol. mp
10
23
6
5
2.2−55.0 °C (decomposition). R (SiO ): 0.55 (EtOAc),
f
2
2
4
stained with KMnO . IR (ATR) ν: 2983, 1359, 1260, 1174,
brownish residue. During heating to 60 °C, a yellowish
solution formed. After 4 h (complete dealkylation to PMPA
detected by HPLC at λ = 254 nm), the solution was cooled in
an ice bath and the pH was adjusted to pH = 2.8−3 using
4
−
1 1
9
65 cm . H NMR, COSY (400 MHz, CDCl ): δ 4.28 (d,
3
1
JP−H = 8.7 Hz, 2H, −P−CH −), 3.12 (s, 3H, −CH ), 1.53 (s,
1
2
3
1
3
8H, −C(CH ) ) ppm. C NMR, HMBC, HSQC (75 MHz,
3
3
2
CDCl ): δ 84.5 (d, J
= 8.5 Hz, −C(CH ) −), 64.2 (d,
conc. NH solution (25%, 0.6 mL, additional water (2−3 mL)
3
C−P
3 3
3
1
3
JC−P = 174 Hz, −P−CH −), 38.2 (−CH ), 30.5 (d, J =
C−P
was added for better stirring). A colorless to slight yellowish
suspension formed, which was stirred for one further hour
while cooling and then stored in a refrigerator overnight. The
suspension was vacuum-filtered, and the filter cake was washed
with ice-cold water (3 × 2 mL) and ice-cold acetone (3 × 2
mL). The filtered solid was first dried on the air and then at 80
2
3
31
3
.9 Hz, −C(CH ) ) ppm. P NMR (162 MHz, CDCl ): δ 7.2
3
3
3
+
ppm. ESI-HRMS: calcd for [M + H] , m/z: 303.1026; found,
m/z: 303.1026.
4
.5. Tenofovir (PMPA), 1. 4.5.1. Variant 1 Using
Mesylate 14. An oven-dried Schlenk flask was charged with
7
96
https://dx.doi.org/10.1021/acs.oprd.0c00473
Org. Process Res. Dev. 2021, 25, 789−798

Organic Process Research & Development
C under high vacuum for 4 h. PMPA was obtained as a
pubs.acs.org/OPRD
Article
°
(determined by Karl-Fischer Titrator). R (C −SiO ): 0.60
f
18
2
colorless powder (1.01 g, 3.52 mmol, 68%).
.6. {[(4-Methylbenzenesulfonyl)oxy]methyl}-
phosphonic Acid, 7. In an oven-dried Schlenk flask, di-tert-
butoxyphosphoryl)methyl 4-methylbenzenesulfonate (13,
.099 g, 0.26 mmol) was heated to 100 °C while purging
(H O). IR (ATR) ν: 3059, 2983, 1698, 1519, 1411, 1153,
2
−
1 1
4
1122, 1056, 1041 cm . H NMR, COSY (400 MHz, D O): δ
2
3
8.41 (s, 1H, H-8), 8.38 (s, 1H, H-2), 4.51 (t, J = 5.0 Hz, 2H,
3
2
(
−NCH −), 3.98 (t, J = 5.0 Hz, 2H, −OCH −), 3.63 (d, J =
2
2
1
3
0
8.7 Hz, 2H, −OCH P−) ppm. C NMR, HMBC, HSQC
(101 MHz, D O): δ 150.2 (C-6), 148.6 (C-4), 145.2 (C-8),
2
with nitrogen. After the solid has melted, it was stirred for
further 15 min and then cooled to r.t. Compound 7 was
obtained as a colorless resin (0.070 g, 0.26 mmol, quant.),
which formed as a colorless solid by trituration. M
2
3
144.8 (C-2), 117.9 (C-5), 70.3 (d, JC−P = 11.6 Hz,
1
−CH O−), 66.9 (d, JC−P = 157 Hz, −OCH P−), 44.0
2
2
3
1
(−NCH −) ppm. P NMR (121 MHz, D O): δ 15.6 ppm.
2
2
+
(
(
C H O PS) = 266.20 g/mol. mp 133.8−134.8 °C. R
ESI-MS m/z: 274.0 (100%, [M + H] ). The spectrometric
8
11
6
f
32
SiO ): 0.19 (EtOAc+10% HOAc), stained with Seebach-
data are consistent with literature values.
2
−
1 1
reagent. IR (ATR) ν: 1365, 1239, 1178, 1026, 939 cm . H
ASSOCIATED CONTENT
sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.oprd.0c00473.
NMR, COSY (400 MHz, DMSO-d ): δ 7.84−7.75 (m, 2H, H-
6
■
2
2
), 7.53−7.45 (m, 2H, H-3), 3.95 (d, J
= 10.0 Hz, 2H,
CH −), 2.43 (s, 3H, −CH ) ppm. C NMR, HMBC,
*
P−H
1
3
−
2
3
HSQC (75 MHz, DMSO-d ): δ 145.4 (C-1), 131.3 (C-4),
6
2
1
2
30.3 (C-3), 128.0 (C-2), 64.0 (d, J
1.2 (-CH ) ppm. P NMR (162 MHz, DMSO-d ): δ 9.5
= 159 Hz, −CH −),
C−P
2
Optimization studies, chromatograms, quantitative
NMR data, PMI, E-factor, atom economy calculations,
and NMR spectra (PDF)
31
3
6
+
ppm. ESI-MS m/z: 266.9 (100%, [M + H] ). The
10
spectrometric data are consistent with literature values.
.7. {[(Methylsulfonyl)oxy]methyl}phosphonic Acid,
1. In an oven-dried Schlenk flask, (di-tert-butoxyphosphoryl)-
4
AUTHOR INFORMATION
Corresponding Author
■
2
methyl methanesulfonate (14, 0.10 g, 0.33 mmol) was heated
to 85 °C while purging with nitrogen. After the solid has
melted, it was stirred for a further 15 min and then cooled to
r.t. 21 was obtained as a reddish oil (0.063 g, 0.33 mmol,
quant.), which formed a colorless to slight reddish solid by
trituration. M (C H O PS) = 190.11 g/mol. mp 94.2−96.5 °C.
Till Opatz − Department of Chemistry, Johannes Gutenberg-
University, Mainz 55128, Germany; orcid.org/0000-
0
002-3266-4050; Email: opatz@uni-mainz.de
Authors
2
7
6
Jule-Philipp Dietz − Department of Chemistry, Johannes
R (C −SiO ): 0.46 (H O + 3% HOAc). IR (ATR) ν: 1365,
f
18
2
2
Gutenberg-University, Mainz 55128, Germany
Dorota Ferenc − Department of Chemistry, Johannes
Gutenberg-University, Mainz 55128, Germany
−
1 1
1
178, 1026, 1011, 939 cm . H NMR, COSY (400 MHz,
DMSO-d ): δ = 4.23 (d, J
s, 3H,−CH ) ppm. C NMR, HMBC, HSQC (101 MHz,
DMSO-d ): δ = 64.2 (d, J
2
= 9.6 Hz, 2H, −P−CH −), 3.22
6
P−H
2
13
(
3
Timothy F. Jamison − Department of Chemistry,
Massachusetts Institute of Technology, Cambridge 02139,
Massachusetts, United States; orcid.org/0000-0002-
1
= 160 Hz,−P−CH −), 36.5 (s,
6
C−P
2
3
1
−
CH ) ppm. P NMR (162 MHz, DMSO-d ): ? = 10.4 ppm.
3
6
+
ESI-HRMS: calcd for [M + H] , m/z: 190.9774; found, m/z:
90.9776.
.8. Adefovir (PMEA), 22. An oven-dried Schlenk flask
8
601-7799
1
B. Frank Gupton − Department of Chemical and Life Sciences
Engineering, Virginia Commonwealth University, Richmond,
Virginia 23284, United States
4
was charged with HEA (23, >98%, 1.00 g, 5.58 mmol, 1.0
equiv) and magnesium tert-butoxide (93%, 3.07 g, 16.7 mmol,
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.oprd.0c00473
3
.0 equiv) under a nitrogen atmosphere. Dry DMA (12 mL)
was added, and the suspension was stirred at 90 °C for 30 min.
While purging with nitrogen, (di-tert-butoxyphosphoryl)-
methyl methanesulfonate (14, 2.53 g, 8.37 mmol, 1.5 equiv)
was added portion-wise within 1 min. The reaction mixture
was stirred at 90 °C for 22 h (conversion of HEA 87% detected
by HPLC at λ = 254 nm). All volatiles were removed in vacuo
at 50−60 °C, and 3 N HCl (10 mL) was added to the orange-
brownish residue. During heating to 60 °C, an orange solution
formed. After 4 h (complete dealkylation to PMEA detected by
HPLC at λ = 254 nm), the solution was cooled in an ice bath
and the pH was adjusted to pH = 2.8−3.0 using NaOH
solution (40 wt %, 4−5 drops). A colorless to slight yellowish
thick suspension formed [additional water (3 mL) was added
for better stirring], which was stirred while cooling for 1 h and
then stored in a refrigerator overnight. The suspension was
vacuum-filtered and washed with ice-cold water (4 × 2 mL)
and ice-cold acetone (3 × 2 mL). The filtered solid was first
dried on the air and then at 80 °C under high vacuum for 3 h
to obtain PMEA [0.98 g, 3.59 mmol, 64%, HPLC-purity (254
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Bill and Melinda Gates
Foundation through the Medicines for All initiative. We thank
all members of the Medicines for All Institute for their
biweekly insights and helpful discussions, Dr J. C. Liermann
(
Mainz) for NMR spectroscopy, and Dr C. J. Kampf (Mainz)
for mass spectrometry.
REFERENCES
■
(
1) Balzarini, J.; Holy, A.; Jindrich, J.; Naesens, L.; Snoeck, R.;
Schols, D.; De Clercq, E. Differential antiherpesvirus and anti-
retrovirus effects of the (S) and (R) enantiomers of acyclic nucleoside
phosphonates: Potent and selective in vitro and in vivo antiretrovirus
activities of (R)-9-(2-phosphonomethoxypropyl)-2,6-diaminopurine.
Antimicrob. Agents Chemother. 1993, 37, 332−338.
3
1
nm): ≥99%]. q- P NMR assay: (99.31 ± 1.29)%. M
(
2) De Clercq, E. Anti-HIV drugs: 25 compounds approved within
(
C H N O P) = 273.19 g/mol. mp 276−278 °C (decom-
25 years after the discovery of HIV. Int. J. Antimicrob. Agents 2009, 33,
307−320.
8
12
5
4
3
2
position) (lit. 282−284 °C ). Water content: 1.1%
7
97
https://dx.doi.org/10.1021/acs.oprd.0c00473
Org. Process Res. Dev. 2021, 25, 789−798

Organic Process Research & Development
pubs.acs.org/OPRD
Article
(
3) De Clercq, E. Tenofovir alafenamide (TAF) as the successor of
tenofovir disoproxil fumarate (TDF). Biochem. Pharmacol. 2016, 119,
−7.
4) Jockusch, S.; Tao, C.; Li, X.; Anderson, T. K.; Chien, M.; Kumar,
(22) Goldwhite, H.; Saunders, B. C. 462. Esters containing
phosphorus. Part XIV. Some tert.-Butyl esters and their reactions. J.
Chem. Soc. 1957, 2409.
1
(
(23) Lapidot, A.; Samuel, D.; Weiss-Broday, M. 127. The synthesis
and hydrolysis of t-butyl phosphate. J. Chem. Soc. 1964, 637.
(24) Mark, V.; Wazer, J. R. V. Tri-t-butyl Phosphite and Some of Its
Reactions. J. Org. Chem. 1964, 29, 1006−1008.
S.; Russo, J. J.; Kirchdoerfer, R. N.; Ju, J. A library of nucleotide
analogues terminate RNA synthesis catalyzed by polymerases of
coronaviruses that cause SARS and COVID-19. Antiviral Res. 2020,
(
25) Grimmond, B. J.; Risehl, M. J.; Roberts, C. R. Contrast
enhancement agents and method of use thereof. Patent US
0140086846 A1, 2014.
26) Du, X.; Eksterowicz, J.; Fantin, V. R.; Jackson, E. L.; Sun, D.;
Ye, Q.; Moore, J.; Zavorotinskaya, T. CD73 inhibitors. Patent WO
019090111 A1, 2019.
27) Kopcho, J. J.; Reddy, K. R.; Matelich, M. C., Ugarkar, B. G.
1
(
80, 104857.
5) del Amo, J.; Polo, R.; Moreno, S.; Díaz, A.; Martínez, E.; Arribas,
J. R.; Jarrín, I.; Hernan, M. A. Incidence and Severity of COVID-19 in
2
(
́
HIV-Positive Persons Receiving Antiretroviral Therapy. Ann. Intern.
Med. 2020, 173, 536−541.
(
2
(
6) https://clinicaltrials.gov/ct2/show/NCT04334928 (accessed
August 30, 2020).
7) Brown Ripin, D. H.; Teager, D. S.; Fortunak, J.; Basha, S. M.;
Process for preparation of cyclic prodrugs of PMEA and PMPA.
Patent US 20030225277 A1, 2003.
(
Bivins, N.; Boddy, C. N.; Byrn, S.; Catlin, K. K.; Houghton, S. R.;
Jagadeesh, S. T.; et al. Process Improvements for the Manufacture of
Tenofovir Disoproxil Fumarate at Commercial Scale. Org. Process Res.
Dev. 2010, 14, 1194−1201.
(28) Becker, M. W.; Chapman, H. H.; Cihlar, T.; Eisenberg, E. J.;
He, G.; Kernan, M. R.; Lee, W. A.; Prisbe, E. J.; Rohloff, J. C.;
Sparacino, M. L. Prodrugs of phosponate nucleotide analoques and
methods for selecting and making same. Patent US 20040018150 A1,
(
8) Arimilli, N. M.; Cundy, C. C.; Dougherty, J. P.; Kim, C. U.;
2
(
004.
Oliyai, R.; Stella, V. J. Antiviral phosphonomethoxy nucleotide
analogs having increased oral bioavailability. Patent US 005922695 A,
29) Holy, A.; Rosenberg, I. Synthesis of 9-(2-
́
phosphonylmethoxyethyl)adenine and related compounds. Collect.
Czech. Chem. Commun. 1987, 52, 2801−2809.
(
1999.
(
9) Riley, D. L.; Walwyn, D. R.; Edlin, C. D. An Improved Process
30) Sowa, S.; Pietrusiewicz, K. M. Chemoselective Reduction of the
for the Preparation of Tenofovir Disoproxil Fumarate. Org. Process
Res. Dev. 2016, 20, 742−750.
(
PO Bond in the Presence of P-O and P-N Bonds in Phosphonate
and Phosphinate Derivatives. Eur. J. Org. Chem. 2019, 2019, 923−938.
10) Derstine, B. P.; Tomlin, J. W.; Peck, C. L.; Dietz, J.-P.; Herrera,
(31) Holy, A.; Dvorakova, H.; Masojídkova, M. Synthesis of
́
̌ ́ ́ ́
B. T.; Cardoso, F. S. P.; Paymode, D. J.; Yue, A. C.; Arduengo, A. J.;
Opatz, T.; et al. An Efficient Synthesis of Tenofovir (PMPA): A Key
Intermediate Leading to Tenofovir-Based HIV Medicines. Org. Process
Res. Dev. 2020, 24, 1420−1427.
Enantiomeric N-(2-Phosphonomethoxypropyl) Derivatives of Purine
and Pyrimidine Bases. II. The Synthon Approach. Collect. Czech.
Chem. Commun. 1995, 60, 1390−1409.
(32) Schwalbe, C. H.; Thomson, W.; Freeman, S. Structural studies
(
11) Arimilli, N. M.; Cundy, C. C.; Dougherty, J. P.; Kim, C. U.;
on bio-active molecules. Part 17. Crystal structure of 9-(2′-
phosphonomethoxyethyl)adenine (PMEA). J. Chem. Soc., Perkin
Trans. 1 1991, 1348−1349.
Oliyai, R.; Stella, V. J. Antiviral phosphonomethoxy nucleotide
analogs having increased oral bioavailability. Patent US RE38333 E,
2003.
(
12) Jansa, P.; Baszczynski, O.; Prochazkova, E.; Dracínsky, M.;
̌
́
́
̌
́
Janeba, Z. Microwave-assisted hydrolysis of phosphonate diesters: An
efficient protocol for the preparation of phosphonic acids. Green
Chem. 2012, 14, 2282.
(
13) Indukuri, V. S. K.; Joga, S. R.; Gorantla, S. R.; Chava, S. Process
for the preparation of tenofovir. Patent US 20140303368 A1, 2014.
14) Elbert, T.; Brehova, P.; Holy, A. The preparation of 3H-labeled
(
̌
́
́
acyclic nucleoside phosphonates and study of their stability. Collect.
Czech. Chem. Commun. 2010, 75, 757−766.
(15) Janeba, Z.; Jansa, P.; Kolman, V.; Baszczynski, O.; Rak, J.;
Ridvan, L.; Kral, V. An improved process for the preparation of
tenofovir disoproxil and pharmaceutically acceptable salts thereof.
Patent WO 2015051874 A1, 2015.
(16) Engelsma, S. B.; Meeuwenoord, N. J.; Overkleeft, H. S.; van der
Marel, G. A.; Filippov, D. V. Combined Phosphoramidite-
Phosphodiester Reagents for the Synthesis of Methylene Bi-
sphosphonates. Angew. Chem., Int. Ed. 2017, 56, 2955−2959.
(17) Clinch, K.; Crump, D. R.; Evans, G. B.; Hazleton, K. Z.; Mason,
J. M.; Schramm, V. L.; Tyler, P. C. Acyclic phosph(on)ate inhibitors
of Plasmodium falciparum hypoxanthine-guanine-xanthine phosphor-
ibosyltransferase. Bioorg. Med. Chem. 2013, 21, 5629−5646.
(18) Kosolapoff, G. M. Preparation of a t-Alkyl Phosphite. J. Am.
Chem. Soc. 1952, 74, 4953.
19) Young, R. W. Di-t-butyl Phosphonate. J. Am. Chem. Soc. 1953,
5, 4620.
20) Gerrard, W.; Isaacs, M. J. D.; Machell, G.; Smith, K. B.; Wyvill,
(
7
(
P. L. 394. Interaction of phosphorus trichloride with alcohols and with
hydroxy-esters. J. Chem. Soc. 1953, 1920.
(21) Cherbuliez, E.; Gandillon, C.; de Picciotto, A.; Rabinowitz, J.
Recherches sur la formation et la transformation des esters XVIII. Sur
la phosphorylation d’alcools tertiaires et sur la vitesse d’hydrolyse des
esters mono-t-alcoylphosphoriques. Helv. Chim. Acta 1959, 42, 2277−
2284.
7
98
https://dx.doi.org/10.1021/acs.oprd.0c00473
Org. Process Res. Dev. 2021, 25, 789−798