Enantioselective Synthesis of Ozanimod, the Active Pharmaceutical Ingredient of a New Drug for Multiple Sclerosis

Article abstract of DOI:10.1002/ejoc.202100058

We report here a short enantioselective synthesis of Ozanimod, a potent modulator of the enzyme Sphingosine-1-phosphate receptor (S1P_R), recently approved by FDA and EMA for the treatment of relapsing-remitting multiple sclerosis. Amongst different synthetic approaches explored, we achieved the best result introducing the stereogenic centre in the last step through imine asymmetric transfer hydrogenation (ATH) using Wills’ catalysts. Besides the reduced numbers of enantiomeric purity controls required, this process culminates in an exceptionally high enantioselective reductive amination obtained with commercially available tethered Ru catalysts. Starting from commercially available 4-cyano-indanone, enantiomerically pure Ozanimod was obtained in 5 steps in 62 % overall yield and 99 % ee.

Full text of DOI:10.1002/ejoc.202100058

Full Papers
doi.org/10.1002/ejoc.202100058
1
2
3
4
5
6
7
8
9
Enantioselective Synthesis of Ozanimod, the Active
Pharmaceutical Ingredient of a New Drug for Multiple
Sclerosis
Claudio Cianferotti,[a] Giuseppe Barreca,[b] Venkatesh Bollabathini,[a] Luca Carcone,[b]
Damian Grainger,[c] Samantha Staniland,[c] and Maurizio Taddei*[a]
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
This paper is dedicated to Professor Franco Cozzi on his 70th Birthday.
We report here a short enantioselective synthesis of Ozanimod,
a potent modulator of the enzyme Sphingosine-1-phosphate
receptor (S1PR), recently approved by FDA and EMA for the
treatment of relapsing-remitting multiple sclerosis. Amongst
different synthetic approaches explored, we achieved the best
result introducing the stereogenic centre in the last step
through imine asymmetric transfer hydrogenation (ATH) using
Wills’ catalysts. Besides the reduced numbers of enantiomeric
purity controls required, this process culminates in an excep-
tionally high enantioselective reductive amination obtained
with commercially available tethered Ru catalysts. Starting from
commercially available 4-cyano-indanone, enantiomerically
pure Ozanimod was obtained in 5 steps in 62% overall yield
and 99% ee.
Introduction
endothelial cells.[4] S1PR has five receptor subtypes, ubiquitous
expressed in multiple organs and systems, and thus involved in
several immune-mediated disorders, including RRMS.[5] Pharma-
cological modulation of S1P receptors can be considered a
suitable therapy for RRMS therapeutic intervention. In early
2020, FDA and EMA approved a potent S1P modulator named
Ozanimod (brand name ZEPOSIA), to treat adult patients with
active RRMS disease. Ozanimod is also under phase III clinical
trials for treatment of ulcerative colitis and Crohn’s disease.[6]
Yet, despite its importance, in the scientific literature, there
is no description of Ozanimod synthesis, most of them located
in patent’s publications.[7–12] The most representative retrosyn-
thetic approach is outlined in Scheme 1. The oxadiazole central
ring is broken into two fragments (A and B) and the chiral 1-
amino-indane fragment B is always obtained from the simplest
4-cyano-1-aminoindane (S)-2 (Scheme 1a).[9]
Multiple Sclerosis (MS) is a chronic inflammatory autoimmune
disease of the central nervous system (CNS), characterized by
demyelination and variable degrees of axonal loss. It is one of
the most common causes of non-traumatic neurological
disability in young adults.[1] Because of the obscure etiology of
MS, finding a cure has been challenging so far.[2] With over
2.5 million patients worldwide, MS is subdivided in almost 4
sub-types, relapsing-remitting multiple sclerosis (RRMS) being
the most populate with approx. 85% of all cases.[3]
The disease is characterized by several neurological dysfunc-
tions dominated by the trafficking and accumulation of
lymphocytes T and B in inflamed tissues. Sphingosine-1-
phosphate (S1P) is a signaling molecule involved in a wide
range of immunological, cardiovascular and neurological proc-
esses, such as the control of lymphocyte trafficking and
regulating heart rate, rhythm and vascular tone. These activities
are mediated through interaction with the sphingosine-1-
phosphate receptor (S1PR), a G-coupled receptor expressed by
dendritic cells, lymphocytes, cardiomyocytes, and vascular
The standard synthesis of this key intermediate depends on
a 4-step process starting from 4-cyano indanone (3) that is
converted into the corresponding tert-butylsulfinyl imine 5
through condensation with chiral auxiliary sulfinamide 4.
°
Diastereoselective reduction at À 78 C with NaBH4, followed by
acid hydrolysis of the auxiliary, gives the key amine in good
yields and good enantiomeric purity (S)-2 (Scheme 1b). Further
elaboration of the ethanolamine arm and oxadiazole condensa-
tion gave the final product 1 in an 8-steps overall synthetic
scheme. To our knowledge, the best complete synthesis of
Ozanimod based on the Ellmann’s chiral auxiliary gave the final
product in 23% overall yield over 8 steps.[10] The process relies
on the use of low temperature requiring cryogenic liquids, use
of dangerous solvents and extensive application of protective
groups. Recently, Gröger and co-workers described an alter-
native route to amine (S)-2 based on enzymatic catalysis.[13]
Transaminase derived from Vibrio fluvialis (VF-TA) worked on
carboxymethyl indanone (7), in the presence of L-Ala as amine
donor, to give the corresponding amino indanone carboxylate
[a] Dr. C. Cianferotti, V. Bollabathini, Prof. M. Taddei
Dipartimento di Biotecnologie, Chimica e Farmacia,
Università degli Studi di Siena
Via A. Moro 2, 53100, Siena, Italy
E-mail: maurizio.taddei@unisi.it
https://docenti.unisi.it/en/taddei-0
[b] G. Barreca, Dr. L. Carcone
Chemessentia srl
Via Bovio 2, 28100; Novara, Italy
[c] Dr. D. Grainger, Dr. S. Staniland
Johnson Matthey Catalysis and Chiral Technologies,
28 Cambridge Science Park, Milton Road,
Cambridge CB4 0FP, UK
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/ejoc.202100058
Part of the “Franco Cozzi’s 70th Birthday” Special Collection.
Eur. J. Org. Chem. 2021, 1924–1930
1924
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100058
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Scheme 2. First synthetic approach based on (S)-CBS oxaborolidine catalytic
enantioselective reduction of ketone 3. DPPA=diphenylphosphoryl azide.
DBU=1,8-Diazabicycloundec-7-ene.
Amidoximes are usually sufficiently stable compounds, and 11
was isolated after chromatography as a viscous waxy com-
pound. A variety of different conditions were then screened for
the synthesis of 1,2,4-oxadiazole ring, both for the activation of
benzoic acid 11 and the aromatization towards the oxadiazole
ring (Table 1).
The original cyclisation proceeded in the presence of EDC
and HOBT for the activation of the carboxylate of compound
12.[18] This is an “atom expensive” reagent to achieve oxadiazole
formation, as a MW of 291.8 is waste just to remove two
molecules of H2O. At first, we tried to activate the benzoic acid
11 as acyl chloride with oxalyl chloride, followed by addition of
Scheme 1. (a) Common retrosynthetic approach to Ozanimod. (b) Synthesis
of key intermediate (S)-2 reported in patents (ref. 10–12). (c) Enzyme
catalyzed enantioselective synthesis of (S)-2 (ref. 13). (S)-2, rac-(2) and 3:
R=CN; 7 and 8: R=À COOMe.
(8) in good conversion and ee. Alternatively, acylation of the
racemic amine rac-(2) with lipase B from Candida antarctica
(CALÀ B) and isopropyl methoxy acetate as the acyl donor, gave
amine (S)-2 with modest overall yield 19%) although with
excellent ee (99%).
Following our interest in optimizing the synthesis of API for
important medicines,[14–16] we tried to design a new approach to
Ozanimod 1 based on catalytic introduction of the stereogenic
center, reduction of unrequired steps and limited use of
protective groups and dangerous solvents.
°
11 and final aromatization promoted by heating to 80 C.
Unfortunately, we observed a very poor reactivity, and most of
the benzoic acid was recovered. Attempts to use, POCl3 or PCl5
to generate the acyl chloride still gave unsatisfactory results.
Therefore, to avoid the use of acyl chloride, we tried a method
involving the use of propyl phosphonic anhydride (T3P®), a
highly reactive cyclic anhydride coupling reagent, with broad
functional group tolerance and easy work-up.[19] Benzoic acid
12, amidoxime 11 and T3P® were mixed in DMF but, after
°
Results and Discussion
heating to 80 C, we did not observe the formation of
oxadiazole 13. Poor results were also obtained with i-BuOCOCl
Due to the relatively simple structure, the general retrosynthesis
was not altered, and the chiral amine (in azide form) was our
first key intermediate. Due to the “phenone” nature of indanone
Table 1. Attempts to 1,2,4-oxadiazole ring formation.
3, it was reduced with (S)-(+)-2-methyl-CBS-oxazaborolidine,
[17]
°
(S)-CBS (10%), and stoichiometric BH3·SMe2 in DCM at À 20 C.
With this procedure, we obtained the alcohol 9 in good yield
and almost complete control of the configuration of the newly
formed stereogenic center after crystallization from ethyl
acetate/hexane (Scheme 2).
Entry
Reaction conditions[a]
13 Yield[b]
The absolute configuration was established transforming
the alcohol 9 into amine (S)-2 through DPPA azidation, that
proceeded with inversion of configuration, and in situ Stau-
dinger reaction with PPh3/H2O. The product of this synthetic
sequence had the same characteristic (NMR/chiral HPLC) of the
amine (S)-2 described in the original patent.[10] Alcohol 9 was
transformed into azide 10 with DPPA and DBU in 70% yield and
then hydroxylamine (free base) was added to the nitrile in
refluxing ethanol to give the hydroxycarbamidoyl derivative 11.
°
1
2
3
4
5
12 +(COCl)2, 5 h, then 11 in DMF at 80 C.
25%
12%
nd
49%
69%
°
12+POCl3 then 11 neat at 80 C
°
12+11+T3P in DMF at 80 C
°
12+CDI then 11 in DMF followed by CDI, 110 C
12+CDI then 11 in DMF followed by TBAF, 110 C.
°
°
[a] Reaction conditions: Acid 12 (0.5 mmol) was mixed at 0 C with the
activation reagent (0.6 mmol) and stirred at rt for 2–8 h. 11 (0.6 mol) was
added in the suitable solvent (1 mL) and the vial heated for 12 h. After
cooling the solvent was evaporated and product 13 isolated by flash
chromatography on silica gel. [b] Isolated yields.
Eur. J. Org. Chem. 2021, 1924–1930
www.eurjoc.org
1925
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100058
of triazine based coupling agents. Thus, 12 was activated with
oxadiazole ring was unstable in presence of such strong bases
and we were never able to obtain compound 17 following
these conditions (Scheme 3).
1
2
3
4
5
6
7
8
9
an equivalent of CDI[20] at 0 C and added to amidoxime 11.
°
After the formation of the linear intermediate (LC-MS analysis),
another equivalent of CDI was added and the mixture was
At this point, the next sequence of events in the retrosyn-
thesis was apparent. For example, alkylation of amine (S)-2 with
bromoethanol THP ether 16 could be possible without affecting
the overall number of synthetic steps of the synthesis. However,
this approach might have also some drawbacks, not least the
tiresome feature of carrying the stereocenter through the
remaining process. We decided therefore to investigate the
possibility to introduce the stereocenter in the last step through
an enantioselective reduction of the imine 18 obtained from
ketone 19 (Scheme 4).
Ketone 19 must be consequently synthesized and, after
oxadiazole retrosynthetic cleavage, amidoxime ketone 20
became the key intermediate (Scheme 4). As the use of NH2OH
required to transform the nitrile into amidoxime is not
compatible with the presence of a free carbonyl, this function
must be protected. We explored two ways: the formation of an
acetal or the reduction to -OH followed by reoxidation after
oxadiazole synthesis. The first approach resulted the more
efficient and only this one is reported here for brevity.
°
heated to 115 C to promote dehydration. After aqueous
workup oxadiazole 13 was isolated in 49% yield. To increase
the yield, the last cyclisation step was attempted using TBAF[21]
and effectively yields in compound 13 were increased to 69%.
Pleased to have found a procedure more efficient than that
reported in the original patent, we approached the elaboration
of the lateral aminoethanol chain. The original report described
the amidation with benzyloxy acetic acid and further amide
reduction and deprotection,[10] another sequence that produced
high amount of wastes, we thus tried to explore the direct
alkylation of the amine.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
The azide of 13 was reduced with catalytic hydrogenation,
and on amine 14 directed alkylation with different bromoetha-
nol derivatives was attempted but the expected product was
never obtained (Scheme 3). Amine 14 was consequently
protected with Boc anhydride in a one-pot reaction, with the
formation of 15 in an unoptimized 60% yield. Unluckily, the use
of NaH to deprotonate the NHBoc group for the further
nucleophilic substitution on 2-bromoethanol THP-ether 16 gave
the required compound 17 in very low yield (12%) together
with a complex mixture of unknown by-products derived from
the cleavage of oxadiazole ring. When other strong bases
(LiHMDS, BuLi) were employed, it was evident that the
Commercially available ketone 3 was then submitted to
different acetalysation attempts using diverse alcohols.
Amongst several trials, we chose 2,2-dimethyl-1,3-propane-
diol 21 that provided 22 as a solid compound, using trimethyl
orthoformate as a dehydrating agent in the presence of a
catalytic amount of PTSA. The optimized yield for acetalysation
(62%) was the lowest of the overall process, but, being at the
first step, the impact on the synthesis efficiency was not too
high (Scheme 5). Acetal 22 was submitted to reaction with
hydroxylamine (free base) in ethanol at rt to give amidoxime 22
in almost quantitative yield. Cyclisation to oxadiazole occurred
°
using our previously explored procedure with CDI at 60–80 C
(Scheme 5). Optimized condition employed only 1.5 eq of CDI
respect to the reagents and TBAF was not required for increase
°
yield, as cyclisation occurred slowly at 80 (12 h) in cyclopentyl
methyl ether (CPME). Without further purification, the acetal of
24 was hydrolyzed with PTSA in wet acetone to give ketone 19
in 93% yield over 2 steps.
The introduction of unprotected ethanolamine 25 on
ketone 19 required a lengthy optimization of reaction con-
ditions. Best results were obtained mixing the two reagents
under acid catalysis and heating gently in benzene with
contemporary evaporation of the azeotropic mixture formed
(rotary evaporator or Dean-Stark apparatus). Unfortunately, the
use of different solvents ((DCM, CPME, THF, EtOH) gave
unsatisfactory results. Although soluble in benzene, compound
19 was insoluble in toluene, thus preventing the use of this
solvent in the reaction. Finally, we found the solution using a
mixture of toluene/ethanol 4/1 v/v in order to obtain a
homogenous solution and prevent the formation of the dieth-
oxy acetal. The toluene/water/ethanol azeotrope was removed
using a rotary evaporator in small scale and a Dean-Stark
apparatus on a larger scale. After a long optimization, imine 18
was obtained as a solid in 96%.
Scheme 3. Attempted Ozanimod via direct alkylation of amine 13 or its Boc
derivative 15.
Scheme 4. New retrosynthetic approach based of enantioselective reduction
of ethanolamine.
Eur. J. Org. Chem. 2021, 1924–1930
www.eurjoc.org
1926
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100058
Table 2. Optimization of enantioselective hydrogenation of imine 18 into
1
2
3
amine 1.[a]
Ent. Catalyst
Solvent
1, Yield [%] ee [%][b]
4
5
6
7
8
9
1
2
3
4
5
6
7
8
C3-[(R,R)-teth-TsDPEN RuCl]
MeOH
2-PrOH
THF
MeCN
DMF
79
57
64
65
63
54 (R)
15
0
44
12
C3-[(R,R)-teth-TsDPEN RuCl]
C3-[(R,R)-teth-TsDPEN RuCl]
C3-[(R,R)-teth-TsDPEN RuCl]
C3-[(R,R)-teth-TsDPEN RuCl]
C3-[(R,R)-teth-TsDPEN RuCl]
C3-[(R,R)-teth-TsDPEN RuCl]
(S,S)-F5DPEN
Toluene 73
2
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
TFE
91
53
23
46
93
86
86
89
90
86
86 (R)
96
10
90
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
9
(S,S)-TsDACH RuCl(p-cym)
(S,S)-MsDPEN RuCl(mesityl.)
C3-[(S,S)-teth-MsDPEN RuCl]
C3-[(S,S)-teth-MtsDPEN RuC]l
C3-[(S,S)-teth-TrisDPEN RuCl]
C4-[(S,S)-teth-MsDPEN RuCl]
C4-[(S,S)-teth-TrisDPEN RuCl]
10
11
12
13
14
15
16
84
99 (S)
99 (S)
68
89
99 (S)
C3-[(S,S)-teth-TrisDPEN RuCl][c] TFE
[a] Reaction conditions: 50 mg of imine 18, catalyst (2 mol%), HCO2H/Et3N
5:2 complex (5 equiv. respect to imine) and 1 mL solvent were added to a
glass vial and the resulting mixture was stirred at 70 C overnight. The vial
was cooled and then diluted with MeOH. A 20 μL aliquot was removed and
added to an HPLC vial containing MeOH (1 mL) and were analyzed by
HPLC (for conditions see SI). [b] Determined via HPLC analysis, see SI. [c]
1 eq of ethanolamine added to the reaction mixture.
°
However, as during the reaction the amount of formic acid
decreases, the corresponding increment of pH promotes the
imine hydrolysis. The same problem occurred when the amount
of formic acid was increased. However, the addition of 1 eq of
ethanolamine to the reaction mixture had a positive effect on
the reaction. In this case, almost full conversion without
decrease of the ee (99% conv, 99% ee) was observed (entry 16
in Table 2).
Based on this encouraging result, we decided to explore the
possibility of a direct reductive amination of ketone 19. There
are scattered examples of direct reductive amination of ketones
under transfer hydrogenation conditions[27] but no general
method is established. The same tethered Ru catalysts used on
imine 19 were explored and, in this case, C3-[(S,S)-teth-TrisDPEN
RuCl] proved to be the most efficient, even lower conversions
were obtained, possibly a result of less favorable reaction pH or
catalysts inhibition.
Scheme 5. Final 5-step enantioselective synthesis of Ozanimod.
Having the imine in hand, we explored the possibility to
carry out the enantioselective hydrogenation asymmetric
hydrogen transfer reaction (ATH)[22–25] a technology widely
applied in industry for the asymmetric reduction of carbonyl
and imine groups. Wills’ ruthenium diamine tethered chiral
catalyst C3-[(R,R)-teth-TsDPEN RuCl][24] was employed at first to
establish the reaction conditions and several solvents were
tested at different temperatures (Table 2). Good conversion in
°
amine 1 was obtained at 70 C using HCOOH/Et3N as the
hydrogen source in almost all solvents, but the best ees were
obtained using MeOH (79% conv., 54% ee, entry 1, Table 2) and
trifluoroethanol, TFE (91% conv., 86% ee, entry 7, Table 2). Then
different tethered and untethered transfer hydrogenation
catalysts were explored under these conditions. From data
collected in Table 2, it is clear that better results were obtained
with tethered catalysts.
In order to force the reaction to completion, the HCOOH/
Et3N mixture was slowly added to the reaction mixture
containing the catalyst and the reagent in TFE at 50 C. After
°
24 h, the Ozanimod amine was obtained in 92% conversion
and 99% ee. The dosing procedure presumably gives improved
conditions for both imine formation and imine reduction, with
better control of reaction pH. Ultimately, reduction of imine 19
with C3-[(S,S)-teth-MtsDPEN RuCl with HCOOH/Et3N (5/2 (5 eq)
plus 1 equivalent of ethanolamine, producing amine 1 in 99%
yield and 99% ee, was the best possible outcome. Pure
Ozanimod was isolated through the formation of the corre-
sponding HCl salt that showed spectral data in accordance with
literature.[10] The determination of ruthenium in the final
product by atomic absorption spectrophotometry showed a
residual Ru contents below 10 ppm.
With the S,S configuration of the catalyst, amine 1 was
formed with the correct stereochemistry. This can be attributed
to the higher stability of the tethered catalysts structure, less
likely to undergo deactivation in the presence of the amine
product. Using C3-[(S,S)-teth-MtsDPEN RuCl] and C3-[(S,S)-teth-
TrisDPEN RuCl][26] 99% ee was reached, together with a good
conversion (84%), although 16% of ketone 19 was formed
(entries 12 and 13 in Table 2). After checking that the ketone
was not present in the starting material, reaction conditions
were modified, finding that a further improvement could be
obtained increasing to some extent the amount of Et3N.
Eur. J. Org. Chem. 2021, 1924–1930
www.eurjoc.org
1927
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100058
silica gel flash chromatography (petroleum ether: ethyl acetate 4:6
Conclusions
1
2
3
4
5
6
7
8
9
(V/V)) to provide 531 mg of the title compound (yield: 79%). 1H
NMR (400 MHz, CDCl3): δ 7.89 (bm, 1H), 7.46 (m, 2H), 7.29 (t, J=7.2,
1H), 5.92 (bs, 1H), 4.85 (bs, 3H), 3.27 (m, 1H), 3.08 (m, 1H), 2.46 (m,
1H), 2.12 (m, 1H) ppm. 13C NMR (101 MHz, CDCl3) δ 147.0, 141.9,
131.4, 128.6, 127.2, 116.6, 108.8, 65.1, 31.4, 29.8 ppm. HRMS (ESI)
calcd for C10H12N5O (M+1)+ 218.1042; found 218.10417.
In conclusion, we developed an enantioselective synthesis of
the RMS drug Ozanimod in just 5 steps in 55% overall yield and
99% ee, an impressive result compared with 23% yield in
8 steps of the patented route. The key step is one of the more
efficient examples of ATH enantioselective reductive amination
described so far, based on commercially available ruthenium
catalysts. This result reiterates that, with the correct choice of
the catalyst, imine hydrogenation is still the privileged access to
enantiomerically pure amines. Use of environmentally sustain-
able conditions without unnecessary loss of chemical materials
and overall shortness are the other strength of the process.
Moreover, the introduction of the stereogenic center at the end
of the synthesis simplify the overall procedure reducing the
analytical efforts required for complete characterization enan-
tiomerically pure intermediates employed for the preparation of
a chiral drug.
(S)-(1-Amino-2,3-dihydro-1H-indene-1-yl)-4-carbonitrile (S)-2. To a
solution of alcohol 9 (0.5 g, 3.1 mmol,) in THF (5 mL) and toluene
(5 mL), diphenylphosphoryl azide (0.104 g, 3.77 mmol) was added
°
at 25 C. DBU (0.62 g, 4.08 mmol) was added monitoring that
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
°
internal temperature does not exceed 25 C. The reaction was
°
maintained under stirring at 25 C until complete conversion into
azide 10 (about 10 hours). Triphenyl phosphine (1.09 g, 4.15 mmol)
was added and, after complete N2 evolution (about 10 hours), water
(75 mL) was added. The reaction was maintained under stirring at
°
50 C until complete conversion (about 15 hours), then a 32% (w/w)
aqueous solution of sodium hydroxide was added up to obtain a
pH >12. The phases were separated, and the organic layer was
washed with water. Formic acid and water were added up to
achieve a pH between 3.5–4, then the phases were separated. The
aqueous phase was washed with toluene. 2-Methyltetrahydrofuran
and a 32% (w/w) aqueous solution of sodium hydroxide were
added to the aqueous phase up to obtain a pH >10. The phases
were separated, and the organic layer was washed with water. The
organic phase was evaporated under reduced pressure and crude
amine was purified by crystallization as the hemitartrate in
methanol (0.76 g, 52% yield). The final product showed the same
properties described in literature.[10]
Experimental Section
(R)-1-Hydroxy-2,3-dihydro-1H-indene-4-carbonitrile 9. (S)-(+)-2-
Methyl-CBS-oxazo-borolidine (3 mL of a 1 M solution in toluene,
3 mmol) and borane-dimethylsulfide (300 μL, 3 mmol) were mixed
into a three-neck round-bottom flask equipped with an addition
funnel under magnetic stirring and nitrogen atmosphere. The
reaction was stirred at rt for 10’ and then dry DCM (50 mL) was
added followed by the remaining part of borane-dimethylsulfide
(5.4 mL, 59.88 mmol in total). The reaction was then cooled to
(S)-5-(3-(1-Azido-2,3-dihydro-1H-inden-4-yl)-1,2,4-oxadiazol-5-yl)-
2-isopropoxybenzonitrile 13. To a solution of 3-cyano-4-isopropox-
ybenzoic acid 12 (132 mg, 0.64 mmol) in DMF (3 ml) maintained at
°
°
À 20 C and through the addition funnel was added dropwise a
0 C, 1,1’-carbonyldiimidazole (115 mg, 0.71 mmol) was added
solution of ketone 3 (4.70 g, 29.94 mmol) over 20’. The reaction was
portionwise. The mixture was maintained under stirring at the same
temperature for about 30 minutes, then a solution of 11 (140 mg,
0.64 mmol) in DMF (2 mL) was added. The mixture was maintained
°
°
stirred at À 20 C for 2 h and then quenched at À 20 C with the
slowly addition of water since complete evolution of gas; the
resulting mixture was then warmed to room temperature and
stirred for additional 30’. The phases were separated, and the
aqueous phase extracted with DCM (3×30 mL). The combined
organic phases were dried over anhydrous sodium sulfate, filtered
and concentrated in vacuo to provide a pale-yellow oil, that was
recrystallized from EtOAc (10 mL) and hexane (50 mL). The precip-
itate was filtered on a Buchner funnel and washed with hexane to
provide compound 9 (3.83 g, 24.1 mmol) as a white solid (yield
80%). MS (ESI) for C9H10NNaO 183 (M+Na)+. 1H NMR (400 MHz,
CDCl3): δ 7.53 (d, J=7.6, 1H), 7.42 (d, J=7.6, 1H), 7.25 (t, J=7.6, 1H),
5.16 (t, J=6.4, 1H), 3.38 (1H, bs, OH), 3.08 (ddd, J=17.2, 8.4, 4.4,
1H), 2.85 (dt, J=16.8, 7.8 Hz, 1H) 2.46 (m, 1H), 1.91 (m, 1H),
ppm [10] 13C NMR (400 MHz, CDCl3): δ 146.8, 146.0, 131.3, 128.5,
127.1, 117.1, 108.5, 75.47, 34.92, 28.97 ppm.
°
under stirring at 25 C for 3 hours, then additional 1,1’ carbon-
yldiimidazole (115 mg, 0.71 mmol) was added. The flask was heated
°
°
to 115 and stirred 24 h at this temperature. After cooling to 25 C,
ethyl acetate (20 mL) was added. The mixture was washed with an
aqueous saturated solution of NaHCO3, 1 NHCl, water and brine.
After drying, the organic layer was evaporated under reduced
pressure and the crude purified by silica gel chromatography
(petroleum ether : ethyl acetate 7:3 (V/ V), yielding 170 mg of the
1
title compound as a colorless oil (yield: 69%). H NMR (400 MHz,
CDCl3): δ 8.30 (d, J=1.6, 1H), 8.24 (dd, J=8.8, 2.0, 1H), 8.07 (d, J=
7.6, 1H), 7.48 (d, J=7.2, 1H), 7.36 (t, J=7.6, 1H), 7.06 (d, J=8.8, 1H),
4.89 (t, J=4.8, 1H), 4.77-4.71 (m, 1H), 3.31 (dddd, J=72.4, 14.4, 8.0,
5.6, 2H), 2.53-2.44 (m, 1H), 2.20-2.11 (m, 1H), 1.43 (d, J=6.0, 6H)
ppm. 13C NMR (400 MHz, CDCl3): δ 172.6, 168.2, 162.3, 143.0, 141.9,
133.5, 133.4, 128.7, 126.9, 126.7, 123.1, 116.3, 114.8, 113.2, 103.4,
72.3, 65.1, 31.7, 31.5, 21.3 ppm. HRMS (ESI) calcd for C21H18N6NaO2
(M+Na)+ 409.1389; found 409.13882.
(S)-1-Azido-N’-hydroxy-2,3-dihydro-1H-indene-4-carboximida-
mide 11 To a solution of alcohol 9 (0.5 g, 3.1 mmol) in THF (5 mL)
and toluene (5 mL), diphenylphosphoryl azide (1.04 g, 3.8 mmol)
was added at 25 C. DBU (0.62 g, 4.1 mmol) was added, monitoring
that internal temperature does not exceed 25 C. The reaction was
°
(S)-tert-Butyl-4-(5-(3-cyano-4-isopropoxyphenyl)-1,2,4-oxadiazol-
3-yl)-2,3-dihydro-1H-inden-1-ylcarbamate 15. Azide 13 (150 mg,
0.39 mmol) was dissolved in EtOH (5 mL) in a round-bottom flask
under magnetic stirring. Pd/C 10% (8.3 mg corresponding to
0.83 mg of Pd, 0.0078 mmol) was added to the solution followed by
Boc2O (94 mg, 0.43 mmol) and the resulting mixture was stirred at
room temperature for 16 h under an atmosphere of H2 (balloon).
The solution was then filtered over celite to remove Pd/C and
evaporated. The crude was dissolved in CH2Cl2 (10 mL) and washed
with HCl 1 N (3×5 mL), water (3×5 mL) and brine (5 mL); the
organic phase was dried over anhydrous sodium sulfate, filtered,
concentrated in vacuo and purified by silica gel flash chromatog-
°
°
maintained under stirring at 25 C until complete conversion (about
10 hours). A 20% (w/w) aqueous solution of sodium carbonate was
added followed by toluene (50 mL). The phases were separated,
and the organic layer was washed with water. The organic phase
was evaporated under reduced pressure providing 0.57 g of crude
azide 10 which was dissolved in ethanol (10 mL). Hydroxylamine
hydrochloride (118 mg, 1.7 mmol) and sodium carbonate (180 mg,
1.7 mmol) were added under stirring and the mixture was refluxed
°
for about 12 hours, then cooled to 25 C and filtered. The filtrate
was evaporated under reduced pressure and the crude purified by
Eur. J. Org. Chem. 2021, 1924–1930
www.eurjoc.org
1928
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100058
raphy (heptane:EtOAc 4:1) to provide compound 15 (108 mg,
0.23 mmol) as a white solid (yield 60%). H NMR (400 MHz, CDCl3,):
J=2.2, 1H), 8.38 (dd, =8.9, 2.2, 1H), 7.96 (dd, J=7.6, 1.1, 1H), 7.60
(tt, J =7.6, 0.8, 1H), 7.16 (d, J=9.0, 1H), 4.88–4.79 (m, 1H), 3.61–3.55
(m, 2H), 2.86–2.80 (m, 2H), 1.51 (d, J=6.1,6H) ppm..[28] MS (ESI) 382
(M+Na)+
1
2
3
4
5
6
7
8
9
1
δ 8.28 (s, 1H), 8.22 (d, J=8.0, 1H), 7.96 (d, J=8.0, 1H), 7.41 (d, J=
7.6, 1H), 7.28 (t, J=7.6, 1H), 7.03 (d, J=9.2, 1H), 5.16 (d, J=7.2, 1H),
4.85 (t, J=6.4, 1H), 4.75–4.69 (m, 1H), 3.38–3.32 (m, 1H), 3.10–3.01
(m, 1H), 2.58–2.56 (m, 1H), 1.82–1.73 (m, 1H), 1.43–1.39 (m, 15H)
ppm. 13C NMR (400 MHz, CDCl3): δ 172.9, 168.8, 162.7, 155.7, 145.3,
142.9, 134.0, 133.8, 128.2, 127.2, 126.8, 123.1, 116.8, 115.2, 113.6,
103.9, 72.7, 55.7, 33.8, 31.5, 28.4, 28.2, 21.7 ppm. HRMS (ESI) calcd
for C26H28N4NaO4 (M+Na)+ 483.2008; found 483.20082.
Ozanimod (1) through hydrogen transfer reduction of imine 18.
To a dispersion of ketone 19 (500 mg, 1.39 mmol) in a 4:1 (V/V)
mixture of toluene and ethanol (60 mL), PTSA (4.7 mg,
0.0247 mmol) and 2-aminoethanol (340 mg, 5.54 mmol) were
added. The mixture was maintained under reflux conditions and
the formed water distilled out using a Dean-Stark condenser
containing activated 4 A molecular sieves. The formation of a
precipitate was observed during the course of the reaction. After
complete conversion (about 12 hours), the mixture was cooled to
5’,5’-Dimethyl-2,3-dihydrospiro[indene-1,2’-[1,3]dioxane]-4-car-
bonitrile 22. To a solution of ketone 3 (30 g, 0.19 mol) in toluene
(200 mL), neopentyl glycol (19.6 g, 0.19 mol) and PTSA (0.72 g,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
°
°
°
0.0038 mol) were added under stirring at 25 C. Trimethyl orthofor-
mate (26.4 g, 0.24 mol) was added and the mixture was maintained
20 C and the solid filtered, washed with toluene and dried at 40 C
under reduced pressure, thus yielding 540 mg of the crude imine
18, yield: 96%; ES/MS 403 (M+H)+, 827, (2 M+Na)+. An aliquot of
50 mg of 18 (0.12 mmol), dispersed in 2,2,2-trifluoroethanol (1 mL),
was mixed with C3-[(S,S)-teth-MtsDPEN RuCl] (1.6 mg,
0.0024 mmol), 2-aminoethanol (7.6 mg, 0.12 mmol) and formic acid
triethylamine complex 5:2 (53.7 mg, 0.62 mmol), under stirring at
°
under stirring at 20–25 C until complete conversion (about
°
12 hours). The reaction was cooled to 5 C, then a 10% aqueous
solution of Na2CO3 (150 mL) and toluene (100 mL) were added. The
resulting phases were separated, and the aqueous layer was
extracted with toluene (80 mL). The collected organic phases were
evaporated under reduced pressure up to obtain a residue which
was triturated in 2-PrOH (130 mL) for 1 hour. The resulting solid
was filtered, washed with 2-propanol and dried at 40 C under
reduced pressure yielding 28.6 g of the title compound (yield:
°
°
25 C. The mixture was maintained under stirring at 50 C until
°
complete conversion (about 24 hours), then it was cooled to 20 C
°
and diluted with methanol. An aliquot of the mixture was analysed
by HPLC according to the method described in literature[10] showing
an enantiomeric excess (ee)=99%. To this methanolic solution a
10% (w/w) solution of hydrogen chloride in methanol (0.13 g,
1
62%). H NMR (300 MHz, DMSO-d6): δ 7.78 (dd, J=25.6, 7.6, 2H),
7.47 (t, J=7.7, 1H), 3.73 (d, J =11.2, 2H), 3.49 (d, J =11.2, 2H), 3.04
(t, J=6.9, 2H), 2.46 (d, J=6.9, 2H), 1.25 (s, 3H), 0.80 (s, 3H) ppm. MS
(ESI) 266 (M+Na)+ C15H17NO2 (243.30): calcd C 74.05, H 7.04, N 5.76;
found C 74.06, H 7.07, N 5.78.
°
0.36 mmol) was added. The mixture was gently heated to 45 C and
maintained under stirring for 2 hours. After cooling, the resulting
solid was filtered, washed with methanol and dried under reduced
pressure to provide 46.5 mg of Ozanimod.HCl (yield 96%) with
chiral HPLC and spectral data in accordance with those reported in
literature.[10]
N’-Hydroxy-5,5-dimethylspiro[1,3-dioxane-2,1’-2,3-dihydroin-
dene]-4’-carboximidamide 23. To a dispersion of hydroxylamine
hydrochloride (20.3 g, 292.2 mmol) in ethanol (490 mL), triethyl-
amine (31.5 g, 311.7 mmol) was added. The mixture was maintained
under stirring at 25 C for 1 hour, then acetal 22 (23.7 g, 97.4 mmol)
was added. The reaction was maintained under stirring at the same
temperature until complete conversion (about 48 hours). The
Ozanimod (1) via reductive amination of ketone 19. To a
dispersion of compound 19 (50 mg, 0.14 mmol) in 2,2,2-trifluoroe-
thanol (1 mL), C3-[(S,S)-teth-TrisDPEN RuCl] (2.0 mg, 0.0028 mmol),
heated to 50 C, 2-aminoethanol (17.0 mg, 0.28 mmol) and formic
acid triethylamine complex 5:2 (60.2 mg, 0.70 mmol) were slowly
added under stirring. The mixture was maintained under stirring at
°
°
°
resulting solid was filtered, washed with ethanol and dried at 40 C
under reduced pressure, yielding 20 g of the title compound
(quantitative yield). An analytical sample was prepared through
recrystallisation with EtOAc. 1H NMR (500 MHz, DMSO-d6): δ 9.55 (s,
1H), 7.47 (dd, J=7.6, 1.2, 1H), 7.41(dd, J=7.6, 1.2, 1H), 7.28 (dd, J=
8.0, 7.3, 1H), 5.72 (s, 2H), 3.71 (d, J=11.1, 2H), 3.47 (dt,J=11.3, 1.1,
2H), 3.02 (t, J=6.9, 2H), 2.34 (dd, J=7.4, 6.5, 2H), 1.26 (s, 3H), 0.78
(s, 3H) ppm. MS (ESI) 275 (MÀ H)À C15H20N2O3 (276.33): calcd C 65.20,
H 7.30, N 10.14; found C 65.17, H 7.29, N 10.12.
°
50 C until complete conversion (about 24 hours), then it was
°
cooled to 20 C and diluted with methanol. An aliquot of the
mixture was analysed by HPLC according to the method described
literature showing an enantiomeric excess (ee)=99%. Purification
through a shorth path of silica with EtOAc/heptane 5 :1, followed
by evaporation of the solvent gave a pure sample of amine 1
1
(45 mg, 90% yield). H NMR (500 MHz, CDCl3) δ=8.44 (s, 1H), 8.36
(dd, J=8.9, 2.2, 1H), 8.10 (d, J=6.7, 1H), 7.57 (d, J=6.4, 1H), 7.41 (t,
J=7.6, 1H), 7.14 (d, J=9.1, 1H), 4.82 (p, J=6.2, 1H), 4.38 (t, J=6.8,
1H), 3.80–3.66 (m, 2H), 3.54–3.42 (m, 1H), 3.28–3.16 (m, 1H), 3.02–
2.89 (m, 2H), 2.61–2.51 (m, 1H), 2.26 (s, 2H), 2.01–1.90 (m, 1H), 1.50
(d, J=6.1, 6H). 13C NMR (126 MHz, CDCl3) δ=173.04, 169.04,
162.76, 146.40, 143.59, 134.15, 133.89, 128.18, 126.99, 123.23,
116.93, 115.31, 113.57, 103.98, 72.76, 62.82, 61.24, 48.44, 33.25,
31.83, 21.76. MS (ESI) 405 (M+H)+
2-Isopropoxy-5-(3-(1-oxo-2,3-dihydro-1H-inden-4-yl)-1,2,4-oxadia-
zol-5-yl)benzonitrile 19. To a dispersion of 3-cyano-4-isopropox-
ybenzoic acid 12 (17 g, 82.8 mmol) in cyclopentyl methyl ether
°
(300 mL) heated to 55 C, 1,1’-carbonyldiimidazole (20.8 g,
128.3 mmol) was added portionwise. The mixture was maintained
under stirring at the same temperature until complete conversion
(about 1 hour), then amidoxime 23 (22.9 g, 82.8 mmol) was added.
The reaction was heated to 80 C and maintained under stirring at
the same temperature until complete conversion (about 12 hours).
After cooling to 60 C, cyclopentyl methyl ether (200 mL) and a
0.5 M aqueous solution of sodium hydroxide (200 mL) were added.
The resulting phases were separated at 50 C and the organic layer
was washed with water (200 mL) at the same temperature. The
organic phase was evaporated under reduced pressure to obtain a
solid which was dispersed in acetone (400 ml) containing PTSA
(1.28 g, 6.73 mmol). The mixture was maintained under stirring at
25 C until complete conversion (about 2 hours), then the resulting
solid was filtered, washed with acetone and dried at 45 C under
°
°
Acknowledgements
°
Support from MIUR (Rome) through the grant Dipartimento di
Eccellenza 2018–2022 is gratefully acknowledged.
°
°
reduced pressure, thus yielding 22.6 g of the title compound (yield:
93%). 1H NMR (500 MHz, CDCl3): δ 8.50 (dd, J=7.6, 1.2, 1H), 8.47 (d,
Eur. J. Org. Chem. 2021, 1924–1930
www.eurjoc.org
1929
© 2021 Wiley-VCH GmbH

Full Papers
doi.org/10.1002/ejoc.202100058
[14] G. Arena, G. Barreca, L. Carcone, E. Cini, G. Marras, H. G. Nedden, M.
Conflict of Interest
1
2
3
4
5
Rasparini, S. Roseblade, A. Russo, M. Taddei, A. Zanotti-Gerosa, Adv.
Synth. Catal. 2013, 355, 1449–1454.
[15] E. Cini, L. Banfi, G. Barreca, L. Carcone, L. Malpezzi, F. Manetti, G. Marras,
M. Rasparini, R. Riva, S. Roseblade, A. Russo, M. Taddei, R. Vitale, A.
Zanotti-Gerosa, Org. Process Res. Dev. 2016, 20, 270–283.
[16] E. Cini, G. Barreca, L. Carcone, F. Manetti, M. Rasparini, M. Taddei, Eur. J.
Org. Chem. 2018, 2622–2628.
The authors declare no conflict of interest.
Keywords: API synthesis
chemistry · Homogenous catalysis
· Asymmetric synthesis · Green
6
7
[17] M. M. Heravi, S. Asadi, N. Nazari, B. M. Lashkariani, Monatsh. Chem. 2016,
147, 961–987.
[18] B. Pipik, G. Ho, J. M. Williams, D. A. Conlon, Synth. Commun. 2004, 34,
1863–1870.
[19] J. K. Augustine, V. Vairaperumal, S. Narasimhan, P. Alagarsamy, A.
Radhakrishnan, Tetrahedron 2009, 65, 9989–9996.
[20] J. M. Ontoria, L. L. Bufi, C. Torrisi, A. Bresciani, C. Giomini, M. Rowley, S.
Serafini, H. Bin, W. Hao, C. Steinkühler, P. Jones, Bioorg. Med. Chem. Lett.
2011, 21, 5274–5282.
[21] A. R. Gangloff, J. Litvak, E. J. Shelton, D. Sperandio, V. R. Wang, K. D. Rice,
Tetrahedron Lett. 2001, 42, 1441–1443.
[22] J. Václavík, P. Šot, B. Vilhanová, J. Pecháček, M. Kuzma, P. Kačer,
Molecules 2013, 18, 6804–6828.
[23] P. Tang, H. Wang, W. Zhang, F.-E. Chen, Green Synth. Catal. 2020, 1, 26–
41.
[24] a) J. Barrios-Rivera, Y. Xu, M. Wills, V. K. Vyas, Org. Chem. Front. 2020, 7,
3312–3342; b) M. Wills, H. G. Nedden, A. Zanotti-Gerosa, Chemical
Records 2016, 16, 2623–2643.
[25] T. C. Nugent, M. El-Shazly, Adv. Synth. Catal. 2010, 352, 753–819.
[26] R. Hodgkinson, V. Jurčík, A. Zanotti-Gerosa, H. G. Nedden, A. Blackaby,
G. J. Clarkson, M. Wills, Organometallics 2014, 33, 5517–5524.
[27] S. D. Boggs, J. D. Cobb, K. S. Gudmundsson, L. A. Jones, R. T. Matsuoka,
A. Millar, D. E. Patterson, V. Samano, M. D. Trone, S. Xie, X. Zhou, Org.
Process Res. Dev. 2007, 11, 539–545.
8
9
[1] C. Baecher-Allan, B. J. Kaskow, H. L. Weiner, Neuron 2018, 97, 742–768.
[2] W.-J. Huang, W.-W. Chen, X. Zhang, Exp. Ther. Med. 2017, 13, 3163–3166.
[3] M. T. Wallin, W. J. Culpepper, E. Nichols, Z. A. Bhutta, T. T. Gebrehiwot,
S. I. Hay, I. A. Khalil, K. J. Krohn, X. Liang, M. Naghavi, A. H. Mokdad, M. R.
Nixon, R. C. Reiner, B. Sartorius, M. Smith, R. Topor-Madry, A. Werdecker,
T. Vos, V. L. Feigin, C. J. L. Murray, Lancet Neurol. 2019, 18, 269–285.
[4] S. J. Taylor, E. H. Demont, J. Gray, N. Deeks, A. Patel, D. Nguyen, M.
Taylor, S. Hood, R. J. Watson, R. A. Bit, F. McClure, H. Ashall, J.
Witherington, J. Med. Chem. 2015, 58, 8236–8256.
[5] A. Huwiler, U. Zangemeister-Wittke, Pharmacol. Ther. 2018, 185, 34–49.
[6] Y. N. Lamb, Drugs 2020, 80, 841–848.
[7] G. Barreca, L. Carcone, M. Taddei, D. M. Grainger, S. L. Staniland, C.
Cianferotti, G. Marras, 2020, WO2020/064818A1.
[8] M. Porcs-Makkay, G. Simig, E. Molnár, J. Pethő, B. Volk, L. Szlávik,
E.Szabó, J. Halász, 2018, WO2018/215807A1.
[9] E. Martinborough, M. F. Boehm, A. R. Yeager, J. Tamiya, L. Huang, E.
Brahmachary, M. Moorjani, G. A. Timony, J. L. Brooks, R. Peach, F. L.
Scott, M. A. Hanson, 2011, WO2011/060392.
[10] E. Martinborough, M. F. Boehm, A. R. Yeager, J. Tamiya, L. Huang, E.
Brahmachary, M. Moorjani, G. A. Timony, J. L. Brooks, R. Peach, F. L.
Scott, M. A. Hanson, 2015, WO 2015/066515A1.
[11] E. Roberts, H. Rosen, S. Brown, M. A. Guerrero, X. Peng, R. Poddutoori,
2009, WO2009/151529.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
[28] P. Turnbull, R. Bakale, M. Marsini, 2020, WO2020/205481, A1.
[12] M. F. Boehm, E. Martinborough, E. Brahmachary, M. Moorjani, J. Tamiya,
L. Huang, A. R. Yeager, 2012, WO2012/158550A2.
[13] F. Uthoff, J. Löwe, C. Harms, K. Donsbach, H. Gröger, J. Org. Chem. 2019,
84, 4856–4866.
Manuscript received: January 19, 2021
Revised manuscript received: March 2, 2021
Accepted manuscript online: March 7, 2021
Eur. J. Org. Chem. 2021, 1924–1930
www.eurjoc.org
1930
© 2021 Wiley-VCH GmbH