Facile synthesis of 2-C-methyl-D-arabino-γ-1,4-lactones and mechanism study

Article abstract of DOI:10.1016/j.tetlet.2018.03.007

In our previous research on the synthesis of 3,5-di-O-benzoyl-2-C-methyl-D-arabino-γ-lactone 4a, the reaction mechanism of DMSO-mediated configuration inversion of C2 hydroxyl group in the lactone was ambiguous. As a follow-up to this research, we used

Full text of DOI:10.1016/j.tetlet.2018.03.007

Accepted Manuscript
Facile synthesis of 2-C-methyl-D-arabino-γ-1,4-lactones and mechanism study
Bingqing Lv, Shuang Guo, Fuqiang Zhu, Tianwen Hu, Qi Guo, Haixia Lin,
Yuanchao Xie, Jingshan Shen
PII:
S0040-4039(18)30304-6
https://doi.org/10.1016/j.tetlet.2018.03.007
TETL 49776
DOI:
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
8 January 2018
27 February 2018
2 March 2018
Please cite this article as: Lv, B., Guo, S., Zhu, F., Hu, T., Guo, Q., Lin, H., Xie, Y., Shen, J., Facile synthesis of 2-
C-methyl-D-arabino-γ-1,4-lactones and mechanism study, Tetrahedron Letters (2018), doi: https://doi.org/10.1016/
j.tetlet.2018.03.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Facile synthesis of 2-C-methyl-D-arabino-γ-1,4-lactones and
mechanism study
Bingqing Lv1, Shuang Guo2, 3, Fuqiang Zhu4, Tianwen Hu4, Qi Guo2, 3, Haixia Lin1*,
Yuanchao Xie2, 3*, Jingshan Shen2, 3
1
Department of Chemistry, Innovative Drug Research Center, College of Sciences, Shanghai
University, 99 Shangda Road, Shanghai 200444, China
2
Shanghai Institute of Materia Medica, Chinese Academy of Sciences, 555 Zuchongzhi Road,
Shanghai 201203, China
3 University of Chinese Academy of Sciences, No. 19A Yuquan Road, Beijing 100049, China
4 Topharman Shanghai Co., Ltd, No.388 Jialilue Road, Zhangjiang Hitech Park, Shanghai 201203
P.R.China
*Address correspondence to this author at Shanghai Institute of Materia Medica, Chinese
Academy of Sciences (CAS), 555 Zuchongzhi Road, Shanghai 201203, P.R. China, 201203. Tel:
+86-20231000-2407, E-mail: xieyuanchao@simm.ac.cn.
Abstract:
In
our
previous
research
on
the
synthesis
of
3,5-di-O-benzoyl-2-C-methyl-D-arabino-γ-lactone 4a, the reaction mechanism of
DMSO-mediated configuration inversion of C2 hydroxyl group in the lactone was
18
ambiguous. As a follow-up to this research, we used O-labelled H2O and DMSO to
investigate the mechanism. It was surprisingly discovered that DMSO served as an
oxygen donor to stereoselectively displace the leaving group, and was regenerated in
the reaction. The protecting groups had no influence on the inversion, but would
significantly affect the yield. We also disclosed another approach for the synthesis of
4a without using the expensive Tf2O, which was more suitable for application.
Key word: nucleoside; 2-C-methyl-D-arabino-γ-lactone; mechanism; SN2 reaction;
18O-labelled DMSO.
2’-subutitued nucleosides which displayed intriguing antiviral and antitumor
activities had attracted strong research attention in recent years.1 2-substituted
2-C-methyl-D-ribono-1,4-lactones were a class of versatile intermediates for such
nucleosides synthesis.2 The most successful anti-HCV drug, sofosbuvir approved in
2013
was
manufactured
from

3,5-di-O-benzoyl-2-deoxy-2-fluoro-2-C-methyl-D-ribono-lactone A,3 and the
2’-chloro analog (IDX21437), currently under phase II clinical trial study was also
prepared from a chloro lactone B,4 shown in Figure 1. In addition, a growing number
of other 2’-subustited nucleosides in clinical or earlier stages of research exhibited
various biological activities and some of them were potentially to be new antiviral or
antitumor agents.5
Figure 1. Anti-HCV nucleosides and their synthetic intermediates
Previously, we reported a simple and efficient route for the synthesis of
intermediates A and B starting from 2-C-methyl-D-ribono-γ-lactone 1, which was a
commercially available material, Scheme 1.6 In this route, the most striking step was
the highly selective and efficient DMSO-mediated α to β inversion of the C2 hydroxyl
group. The reaction went very fast and easily gave the product in at least 80% yield
by crystallization. However, the inversion mechanism had not been studied. In order
to find out the source of the oxygen atom of C2-β-OH, herein we investigated the
reactions using 18O-labelled H2O and DMSO, respectively. A series of
2-C-methyl-D-ribono-γ-lactones with different protecting groups was also synthesized.
With these substrates, we further proved that the reaction didn’t rely on the C3 and C5
protecting groups.
Scheme 1. our previously reported route for the synthesis of A and B
Results and Discussion
Since the C2-α-OH of intermediate 2a was activated to be a good leaving group
(OTf), neighboring group participation may occur if the leaving group departed. At
the beginning, we speculated that H2O was the source of the C2-β-OH of the product.
To validate our assumption, the reaction was carried carefully in the presence of
18O-labelled H2O. However, the result did not agree well with our original idea. The
content rate of 18O in 4a was no more than 10% based on ESI-MS analyses (Figure S1

in supporting information), indicating that the reaction proceeded not mainly through
this way, Figure 2.
Figure 2. The original proposed mechanism of H2O-mediated inversion
Then, we turned to investigate if DMSO took part in the reaction. 18O-labelled
DMSO could be synthesized from dimethylsulfur dibromide and [18O] H2O (97
atom%18O) in the presence of triethylamine.7 The purity of prepared [18O] DMSO was
about 81% roughly judged from the EI-MS. After [18O] DMSO was added into a
dichloromethane solution of 3a, it was curiously found that the reaction went much
slower, and was completed over 12 hours at room temperature. Not too surprisingly,
the [18O] was successfully incorporated in the product as well as the debenzoylated
product with a high ratio of 74% (Figure S4 and S5 in supporting information). The
result convincingly proved that the oxygen atom of C2-β-OH came from DMSO. In
the reaction, DMSO likely served as a nucleophile that replaced the leaving group
(OTf) in a SN2 manner. The transition state was highly unstable, and easily
hydrolyzed to afford the C2-β-OH lactone and regenerate DMSO, Figure 3. In an
early research, synthesis of bromohydrin using N-bromosuccinimide (NBS) and
DMSO shared a similar mechanism with the present reaction.8 This combination was
also welcomed for the synthesis of α-acyloxy ketones and α-amino carbonyl
compounds.9 Swern oxidation was well-known, but Technically, DMSO did not
participate in the reaction directly.10 Indeed, by now there were almost no other cases
of demonstrating DMSO or sulfoxide as an oxygen donor in substitution reaction,
except a recent study of Zhang’s group on the gold-catalyzed intramolecular redox
reaction of sulfinyl alkynes.11
Figure 3. The possible mechanism of DMSO-mediated inversion
From the mechanism, this inversion reaction seemed to be independent of the
protecting groups at C3-OH and C5-OH positions. In confirmation of it, we prepared
a series of lactones which had different protecting groups at the two positions,

Scheme 2. Following the same procedure described in Scheme 16, all products were
successfully obtained with inverted C2 hydroxyl groups, shown in Table 1. For the
acyl groups, the yield was similar among compounds 4a-4e, and a little lower for
compounds 4f-4g. Notably, the inversion also occurred in the TBS and TIPDS
containing substrates (Entry 8 and Entry 9), which strongly supported the proposed
mechanism. In Entry 8-10, triflation of 2h-2j did not go to completion, and the
products 3h-3j were found very labile during purification by column chromatography.
As a result, compounds 4h-4j were obtained in low yields. In contrast, the aroyl
protecting triflates were much more stable, such as 3a which could react with
different nucleophiles (tetrabutyl ammonium bromide, aqueous ammonia, sodium
azide) in acetone to give 2-α-methyl-β-substituted arabino-1,4-lactones 4k-4m. The
two compounds (4k, X = Br and 4m, X = N3) were obtained smoothly in about 80%
yields with no need of chromatographical purification. As for 4l (X = NH2), the yield
was relatively lower (30%), which was largely due to lack of stability of the lactone
ring or Bz protecting groups under basic condition. Such lactones were useful
building blocks in the synthesis of novel 2’-substituted nucleosides.
Scheme 2. Synthesis of lactones 4a-4j and 4k-4m
Table 1, the results of DMSO-mediated C2-OH inversion for different substrates.
Entry Product
R1
R2
yielda
73%
84%
79%
77%
80%
45%
1
2
4a
4b
4c
4d
4e
4f
Bz
Bz
4-Cl-Bz
4-Br-Bz
3-MeO-Bz
4-Me-Bz
Pivaloyl
4-Cl-Bz
4-Br-Bz
3-MeO-Bz
4-Me-Bz
Pivaloyl
3
4
5
6
7
4g
4h
4i
furan-2-carbonyl furan-2-carbonyl 53%
8
TIPDS
10%
15%
15%
9
TBS
Trt
TBS
Bz
10
4j
a: The yield was calculated from 2

With the collecting results, the DMSO-mediated inversion mechanism became
18
much clear. To further validate it, we performed the reactions with O-labelled H2O
(5 eq) and unlabelled DMSO (2 eq) again, and specifically analyzed the DMSO. It
was found that [18O] DMSO was produced obviously, accounting for nearly 40% in
the mixture (Figure S6 in supporting information). This result agreed well with the
proposed mechanism, Figure 4. The newly generated [18O] DMSO would also take
part in the reaction, so it was reasonable to observe a small amount of 18O-labelled 4a
in the aforementioned reaction using 18O-labelled H2O.
Figure 4. The mechanistic course revealed by the generated DMSO
In our previous study, the DMSO-mediated conversion of C2-OH was achieved
only through the triflate 3a, instead of the mesylate 3n or tosylate 3o even under high
temperature. This could be well explained on account of the significant difference in
the leaving ability between OTf and OMs or OTs. However, because
trifluoromethanesulfonic anhydride was expensive and not atom-economic, from an
application perspective, this route was not suitable for large-scale production of
pharmaceutical intermediates A or B.
It was important to note that the leaving groups of 3n or 3o were located at the
α-position of the lactone, so it was feasible to make the conversion by some
nucleophilic reagent. After many attempts, we found that reaction of the mesylate 3n
with KNO2 or NaNO2 also gave the desired product, but no reaction was detected for
3o, Scheme 3. We had screened the reaction conditions, such as the solvent,
temperature, ratio of reagents and catalysts. Finally, DMSO was found to be the
preferred solvent, and the result was better with KNO2 relative to NaNO2 in aspect of
reaction time and impurities. In addition, catalytic amount of Bu4NI would improve
the yield to 70%, but still lower compared with the former approach.
Scheme 3. A new route for the synthesis of compound 4a
Conclusion

In the present research, using 18O-labelled H2O and DMSO, we determined that
DMSO acted as a nucleophile and played a crucial role in the synthesis of
2-C-methyl-D-arabino-γ-lactones. The mechanism was special because DMSO
participated in the reaction directly. We also synthesized some 2-substituted
2-C-methyl-D-arabino-1,4-lactones, which may be useful for the construction of
novel nucleosides. In addition, a new approach for the preparation of 4a avoid using
the expensive Tf2O was disclosed, and potentially to be applied for the scalable
synthesis.
Acknowledgment. This work was supported by National Science Foundation for
Young Scientists of China (Grant No. 21502209). The authors gratefully
acknowledged Topharman Shanghai Co., Ltd for collaboration.
Supplementary data
1
Supplementary data (experimental procedures, spectral data, copies of H-NMR,
13C-NMR spectra) can be found in the supporting information.
References
1. (a) Jordheim LP, Durantel D, Zoulim F, Dumontet C. Nat Rev Drug Discov.
2013;12:447-464;
(b) Cavaliere A, Probst KC, Westwell AD, Slusarczyk M. Future Med Chem.
2017;9:1809-1833;
(c) Xie Y, Ogah CA, Jiang X, Li J, Shen J. Curr Drug Targets. 2016;17:1560-1576.
2. (a) Pierra C, Benzaria S, Amador A, Moussa A, Mathieu S, Storer R, Gosselin G.
NUCLEOS NUCLEOT NUCL. 2005;24:767-770;
(b) Eldrup AB, Prhavc M, Brooks J, Bhat B, Prakash TP, Song Q, Bera S, Bhat N,
Dande P, Cook PD, Bennett CF, Carroll SS, Ball RG, Bosserman M, Burlein C,
Colwell LF, Fay JF, Flores OA, Getty K, LaFemina RL, Leone J, MacCoss M,
McMasters DR, Tomassini JE, Von Langen D, Wolanski B, Olsen DB. J Med
Chem. 2004;47:5284-5297;
(c) Reddy PG, Chun BK, Zhang HR, Rachakonda S, Ross BS, Sofia MJ. J Org
Chem. 2011;76:3782-3790;
(d) Girijavallabhan V, Bogen S, Pan W, Dang Q. U.S. Patent 9242988B2, 2016.
3. (a) Sofia MJ, Bao D, Chang W, Du J, Nagarathnam D, Rachakonda S, Reddy PG,
Ross BS, Wang P, Zhang HR, Bansal S, Espiritu C, Keilman M, Lam AM, Steuer
HM, Niu C, Otto MJ, Furman PA. J Med Chem. 2010;53:7202-7218;
(b) Rodriguez-Torres M. Expert Rev Anti Infect Ther. 2013;11:1269-1279;
(c) Li Z, Ding H, Yan J, Dai Y, Gu W. PCT Patent WO2016112746A1, 2016.
4. Alexandre FR, Badaroux E, Bilello JP, Bot S, Bouisset T, Brandt G, Cappelle S,
Chapron C, Chaves D, Convard T, Counor C, Da Costa D, Dukhan D, Gay M,
Gosselin G, Griffon JF, Gupta K, Hernandez-Santiago B, La Colla M, Lioure MP,

Milhau J, Paparin JL, Peyronnet J, Parsy C, Pierra Rouviere C, Rahali H, Rahali R,
Salanson A, Seifer M, Serra I, Standring D, Surleraux D, Dousson CB. Biorg Med
Chem Lett. 2017;27:4323-4330.
5. (a) Vere Hodge RA. Antiviral Res. 2017;137:23-40;
(b) Yang Q, Kang J, Zheng L, Wang XJ, Wan N, Wu J, Qiao Y, Niu P, Wang SQ,
Peng Y, Wang Q, Yu W, Chang J. J Med Chem. 2015;58:3693-3703;
(c) Norkin M, Richards AI. Expert Rev Anticancer Ther. 2015;15:1261-1266;
(d) Chang W, Bao D, Chun BK, Naduthambi D, Nagarathnam D, Rachakonda S,
Reddy PG, Ross BS, Zhang HR, Bansal S, Espiritu CL, Keilman M, Lam AM, Niu
C, Steuer HM, Furman PA, Otto MJ, Sofia MJ. ACS Med Chem Lett.
2011;2:130-135.
6. Xie Y, Zhang J, Tian G, Xu M, Hu T, Jiang X, Shen J. Tetrahedron Lett.
2015;56:4345-4348.
7. Fenselau AH, Moffatt JG. J Am Chem Soc. 1966;88:1762-1765.
8. Dalton DR, Dutta VP, Jones DG. J. Am Chem Soc. 1968;90:5498-5501.
9. Prasad PK, Reddi RN, Sudalai A. Org Lett. 2016;18:500-503.
10. (a) Torssell K. Tetrahedron Letters. 1966;7:4445-4451;
(b) Moffatt JG. J Org Chem. 1971;36:1909-1912.
11. (a) Li G, Zhang L. Angew Chem Int Ed. 2007;46:5156-5159;
(b) Zhang L. Acc Chem Res. 2014;47:877-888.

 A DMSO-participating reaction mechanism for the synthesis of
2-C-methyl-D-arabino-γ-1,4-lactones
 The mechanism relying on the C2 leaving groups, but not the C3 or C5 protecting
groups
 A novel
approach
suitable
for
the
scalable
preparation
of
3,5-di-O-benzoyl-2-C-methyl-D-arabino-γ-lactone 4a