A High-Yielding Synthesis of EIDD-2801 from Uridine**

Article abstract of DOI:10.1002/ejoc.202001340

A simple reordering of the reaction sequence allowed the improved synthesis of EIDD-2801, an antiviral drug with promising activity against the SARS-CoV-2 virus, starting from uridine. Compared to the original route, the yield was enhanced from 17 % to 61 %, and fewer isolation/purification steps were needed. In addition, a continuous flow procedure for the final acetonide deprotection was developed, which proved to be favorable toward selectivity and reproducibility.

Full text of DOI:10.1002/ejoc.202001340

European Journal of Organic Chemistry
Accepted Article
Accepted Article
Title: A High-Yielding Synthesis of EIDD-2801 from Uridine
Authors: Alexander Steiner, Desiree Znidar, Sándor B. Ötvös, David R.
Snead, Doris Dallinger, and C. Oliver Kappe
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.202001340
Link to VoR: https://doi.org/10.1002/ejoc.202001340
01/2020

10.1002/ejoc.202001340
European Journal of Organic Chemistry
COMMUNICATION
A High-Yielding Synthesis of EIDD-2801 from Uridine
Alexander Steiner,[a,b] Desiree Znidar,[a,b] Sándor B. Ötvös,[a,b] David R. Snead,[c] Doris Dallinger,*[a,b]
and C. Oliver Kappe*[a,b]
[a]
A. Steiner, D. Znidar, Dr. S. B. Ötvös, Dr. D. Dallinger and Prof. Dr. C. O. Kappe
Institute of Chemistry
University of Graz, NAWI Graz
Heinrichstrasse 28, 8010 Graz, Austria
E-mail: do.dallinger@uni-graz.at, oliver.kappe@uni-graz.at
http://goflow.at
[b]
[c]
A. Steiner, D. Znidar, Dr. S. B. Ötvös, Dr. D. Dallinger and Prof. Dr. C. O. Kappe
Center for Continuous Flow Synthesis and Processing (CCFLOW)
Research Center Pharmaceutical Engineering GmbH (RCPE)
Inffeldgasse 13, 8010 Graz, Austria
Dr. D. R. Snead
Medicines for All Institute
737 N. 5th St., Box 980100,
Richmond, Virginia 23298-0100
Supporting information for this article is given via a link at the end of the document.
Abstract: A simple reordering of the reaction sequence allowed the
improved synthesis of EIDD-2801, an antiviral with promising activity
against the SARS-CoV-2 virus, starting from uridine. Compared to the
original route, the yield was enhanced from 17% to 61%, and fewer
isolation/purification steps were needed. In addition, a continuous flow
procedure for the final acetonide deprotection was developed, which
proved to be favorable toward selectivity and reproducibility.
of a pill, the potential treatment against COVID-19 would be made
more accessible compared to remdesivir that currently has to be
administered intravenously.
EIDD-2801 was developed by Emory University researchers,
and the disclosed route consists of a five-step synthesis starting
from uridine (Scheme 1a).[5] This was the only reported synthesis
in the open literature until recently, when routes toward EIDD-
2801 were reported that replaced uridine for cytidine.[6,7] In the
original route, first an acetonide protection followed by selective
esterification of the 5’-hydroxy group is performed in a one-pot
fashion. Then, the molecule is activated by introduction of the
1,2,4-triazole moiety which is further displaced by hydroxylamine.
After removal of the acetonide protecting group, EIDD-2801 is
obtained via a complex crystallization procedure. As no isolated
yield for the final deprotection step has been disclosed in the
patent, an overall yield (17%) can only be stated for the synthesis
of protected EIDD-2801. The main disadvantage of the reported
route is the low yield (29%) in the triazole coupling step.
The current medical crisis caused by the SARS-CoV-2 virus, has
impacted people’s lives and economy worldwide, with presently
more than 35 million confirmed cases.[1] In view of the severity of
the COVID-19 pandemic, there is a continuous quest for new
drugs for the treatment of this disease. EIDD-2801, a NHC (-D-
N4-hydroxycytidine) prodrug, has shown potent activity against
multiple CoVs in animal studies[2] and is currently in Phase II
clinical trials on symptomatic patients with COVID-19.[3] This
antiviral candidate is structurally similar to remdesivir, but blocks
RNA polymerase differently, which possibly makes them
complimentary.[4] Since EIDD-2801 can be taken orally in the form
In 1997, Reese and co-workers reported the triazolation of
Scheme 1. Patented route (a) versus proposed route (b) toward EIDD-2801.
1
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001340
European Journal of Organic Chemistry
COMMUNICATION
uridine to proceed in 89% yield after a simple extractive work-up
and crystallization.[8] Therefore, we anticipated, that a simple
reordering of the steps in the original synthetic sequence would
result in an overall higher yield and improved isolation procedures
Further, by starting from uridine the impurity profile should closely
match that of the commercial process which might accelerate
uptake of the modified process by minimizing regulatory changes.
As illustrated in Scheme 1b, we planned our synthesis toward
EIDD-2801 starting with the triazolation according to Reese and
co-workers. The other transformations (acetonide protection,
esterification, hydroxyamination) should ideally proceed in a
similar approach as in the patented route. Acetonide deprotection
from 4 as final step and isolation of EIDD-2801 is envisaged as
reported in the original patent.[5]
distillation step of 1.2 volumes proved to be less efficient, as 1.3
equiv. of anhydride were required. Acetonide ester 3 was
obtained in quantitative yield and ≥99% HPLC purity after
extractive work-up in this one-pot procedure.
Following the original procedure of Reese and co-workers
which involves a one-pot TMS protection of uridine and
phosphorus oxychloride-promoted triazole coupling using Et3N as
base,[8] 1 was obtained in 74% yield. When changing the base to
N-methylpyrrolidine the yield of 1 increased to 88% (Scheme 2).
Product isolation consisted of a simple filtration/washing step, as
1 precipitated during deprotection from the MeOH/AcOH mixture.
Additionally, we attempted to reduce the amount of 1,2,4-triazole
(10 equiv.), because of the associated health risks.[9]
Unfortunately, less triazole also leads to lower isolated yields of 1
(8 equiv.: 78%, 6 equiv.: 58%). Compared to the original route,[5]
we were able to increase the yield for the triazolation step from
29% to 88% by introducing the triazole moiety as first step.
Importantly, also column chromatographic work-up could be
avoided.
Scheme 3. Steps 2 and 3: One-pot acetonide protection and esterification. See
the Supporting Information for more details.
Acetonide ester 3 can be readily converted to hydroxylamine 4,
by stirring in i-PrOH at room temperature with 1.5 equiv. of
hydroxylamine (50 wt% in H2O) for 20 min (Scheme 4). After
extractive work-up, 4 was isolated in 90% yield and 91% HPLC
purity. The purity could be improved to 99% by washing with cold
diethyl ether, but ca 50% of product was lost in this washing step.
Therefore, the washing step was omitted, and the crude product
after extraction was used for the acetonide deprotection step.
Scheme 4. Step 4: Hydroxylamine formation.
Scheme 2. Step 1: Triazolation of uridine.
Removal of the acetonide protecting group of 4 leading to EIDD-
2801 has been described to proceed in formic acid at room
temperature overnight (see Scheme 1a).[5] However, by
performing the reaction exactly under the reported conditions only
a 47% conversion to EIDD-2801 was obtained. Heating to 60 °C
for 5 h or 100 °C for 30 min resulted in the increased formation of
side products. An acid screening (see Table S2) revealed that
H2SO4 in combination with i-PrOH as solvent furnished the
highest conversion (80%) to EIDD-2801 after heating at 60 °C for
30 min. The main side reaction was found to be the ester
hydrolysis yielding N-hydroxycytidine 5 (EIDD-1931, see also
Figure 1), in particular when the reaction was conducted at higher
temperatures.
Since the same solvent system (i-PrOH) was employed in the
hydroxyamination and acetonide deprotection, we envisioned that
these two steps could be combined to a one-pot procedure. A
similar process has been recently demonstrated in an alternative
route toward EIDD-2801.[7] During batch optimization studies (see
Table S3) we experienced reproducibility issues that were mainly
related to the exothermic nature upon H2SO4 addition in the
deprotection step. Therefore, we decided to transfer the batch
process to continuous flow, where better mixing and temperature
To selectively introduce the isobutyl ester at the 5’-hydroxy
position, the other two hydroxy groups need to be protected first.
When using the conditions for the acetonide protection of uridine
described in the original patent (5 mol% H2SO4 in acetone),[5] no
conversion of 1 to the desired acetonide 2 was observed. Higher
concentrations (0.5 equiv.) of H2SO4 lead to hydrolysis of the
triazole moiety. Addition of 2 equiv. of 2,2-dimethoxypropane
(DMP) did not significantly improve the reaction. However, when
exchanging acetone for MeCN as solvent in combination with the
use of DMP, a 73% conversion to 2 was achieved. Further
optimization revealed, that the amount of H2SO4 could be reduced
to 5 mol% (Scheme 3).
Upon attempting the acetonide formation and esterification in
a one-pot fashion similar to the patented route (see Scheme 1),
we observed that 3 equiv. of isobutyric anhydride were needed in
order to reach full conversion of 2. The excess of anhydride was
necessary because it was quenched by the 2 equiv. of MeOH that
were released from DMP. Therefore, after stirring of 1 with DMP
and H2SO4 in MeCN for 30 min at room temperature, a stepwise
azeotropic distillation was performed to remove MeOH. Applying
this protocol resulted in only 1.1 equiv. of isobutyric anhydride
being necessary for full conversion of 2 (Scheme 3). A single
2
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001340
European Journal of Organic Chemistry
COMMUNICATION
control allows for superior control of exotherms, and in return
would provide reproducible results.[10–12]
A simple flow set-up was therefore established using a heated
reaction coil together with individual streams for introducing
solutions of the substrate and the acid reagent, which were
combined in a Y-shaped mixer (Figure 1A). Triazole 3 was
converted to hydroxylamine 4 in batch under the same conditions
as shown in Scheme 5. However, 4 was not isolated, but after 15
min of stirring, the reaction mixture was introduced directly into
the flow reactor via a sample loop for the acetonide deprotection.
The flow process was investigated in various solvents using
HCOOH, H2SO4, CF3COOH or TfOH as acid component (Tables
S4–S7). During the deprotection optimization studies, ester
hydrolysis and hydroxyl amine–hydroxyl exchange occurred as
most prominent side reactions yielding compounds 5 and 6
(Figure 1B, see also the Supporting Information). The effects of
reaction temperature, residence time and excess of acid were
therefore carefully examined with the aim to minimize side product
formation while simultaneously maximize conversion to EIDD-
2801 (see Tables S4–S7 for details). Similarly as in batch, H2SO4
proved to be superior to the other tested acids (see Table S5).
Because of precipitation issues in i-PrOH, MeOH was selected as
solvent. The highest conversion to EIDD-2801 (79%) was
obtained at only 5 min of residence time using 2.75 equiv. of
H2SO4 at 100 °C, while 5 (11%) and 6 (8%) were at a minimum.
Importantly, the flow process under optimum conditions not only
permitted higher yields, but it also enabled a significant chemical
intensification and a proper reproducibility as compared to our
initial batch attempts. By employing these flow conditions for
scale-out (16 min), pure EIDD-2801 was obtained in 69% isolated
yield (307 mg) after chromatographic purification.
Scheme 5. Improved protocol for the synthesis of EIDD-2801.
Acknowledgements
We thank the Bill and Melinda Gates Foundation for their
longstanding support of our research.
Keywords: EIDD-2801 • COVID-19 • continuous flow •
acetonide deprotection • triazolation
[1]
[2]
https://www.who.int/emergencies/diseases/novel-coronavirus-2019
(accessed October 7, 2020).
T. P. Sheahan, A. C. Sims, S. Zhou, R. L. Graham, A. J. Pruijssers,
M. L. Agostini, S. R. Leist, A. Schäfer, K. H. Dinnon, L. J. Stevens,
J. D. Chappell, X. Lu, T. M. Hughes, A. S. George, C. S. Hill, S. A.
Montgomery, A. J. Brown, G. R. Bluemling, M. G. Natchus, M.
Saindane, A. A. Kolykhalov, G. Painter, J. Harcourt, A. Tamin, N. J.
Thornburg, R. Swanstrom, M. R. Denison, R. S. Baric, Sci. Transl.
Med. 2020, 12, 5883.
[3]
[4]
https://www.clinicaltrials.gov/ct2/show/NCT04405570?intr=EIDD-
2801&draw=2 (accessed October 7, 2020).
https://www.trialsitenews.com/merck-positioning-to-take-the-lead-u-
s-covid-19-antiviral-market-from-remdesivir-eidd-2801/ (accessed
October 7, 2020).
[5]
[6]
G. R. Painter, G. R. Bluemling, M. G. Natchus, D. Guthrie, N4-
Hydroxy Cytidine and Derivatives and Anti-Viral Uses Related
Thereto, 2019, WO2019113462A1.
N. Vasudevan, G. P. Ahlqvist, C. P. McGeough, D. J. Paymode, F.
S. P. Cardoso, T. Lucas, J.-P. Dietz, T. Opatz, T. F. Jamison, B. F.
Gupton, D. Snead, Chem. Commun. 2020, DOI
10.1039/D0CC05944G.
[7]
V. Gopalsamuthiram, C. Williams, J. Noble, T. F. Jamison, B. F.
Gupton, D. R. Snead, Synlett 2020, DOI 10.1055/a-1275-2848.
A. Miah, C. B. Reese, Q. Song, Nucleosides and Nucleotides 1997,
16, 53–65.
[8]
Figure 1. Steps 4 and 5: Telescoped batch hydroxyamination and continuous
flow acetonide deprotection. (A) Schematic representation of the reaction set-
up. (B) Structures of identified side products.
[9]
1,2,4-Triazole is a potential CMR substance, as it is suspected to be
reprotoxic.
[10]
[11]
[12]
S. R. L. Gobert, S. Kuhn, L. Braeken, L. C. J. Thomassen, Org.
Process Res. Dev. 2017, 21, 531–542.
In conclusion, an improved protocol was developed to access
EIDD-2801 from uridine as starting material (Scheme 5). By
strategic reordering of synthetic steps and by employing a
continuous flow process for the final acetonide deprotection, the
overall yield was improved from 17% to 61%. Importantly, this
strategy presents fewer and simplified product isolation
procedures, as two telescoped procedures (acetonide
B. Gutmann, D. Cantillo, C. O. Kappe, Angew. Chemie - Int. Ed.
2015, 54, 6688–6728.
M. B. Plutschack, B. Pieber, K. Gilmore, P. H. Seeberger, Chem.
Rev. 2017, 117, 11796–11893.
protection/esterification
and
hydroxyamination/acetonide
deprotection) are included within the 5-step route.
3
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202001340
European Journal of Organic Chemistry
COMMUNICATION
Entry for the Table of Contents
An improved protocol was developed for the synthesis of EIDD-2801, which is currently in Phase II clinical trials for the treatment of
COVID-19. The process employs uridine as starting material and proceeds via triazolation, one-pot acetonide protection/esterification
and telescoped hydroxyamination/continuous flow acetonide deprotection.
4
This article is protected by copyright. All rights reserved.