Investigation of the effect of different linker chemotypes on the inhibition of histone deacetylases (HDACs)

Article abstract of DOI:10.1016/j.bioorg.2020.104462

Histone Deacetylases (HDACs) are among the most attractive and interesting targets in anticancer drug discovery. The clinical relevance of HDAC inhibitors (HDACIs) is testified by four FDA-approved drugs for cancer treatment. However, one of the main drawbacks of these drugs resides in the lack of selectivity against the different HDAC isoforms, resulting in severe side effects. Thus, the identification of selective HDACIs represents an exciting challenge for medicinal chemists. HDACIs are composed of a cap group, a linker region, and a metal-binding group interacting with the catalytic zinc ion. While the cap group has been extensively investigated, less information is available about the effect of the linker on isoform selectivity. To this aim, in this work, we explored novel linker chemotypes to direct isoform selectivity. A small library of 25 hydroxamic acids with hitherto unexplored linker chemotypes was prepared. In vitro tests demonstrated that, depending on the linker type, some candidates selectively inhibit HDAC1 over HDAC6 isoform or vice versa. Docking calculations were performed to rationalize the effect of the novel linker chemotypes on biologic activity. Moreover, four compounds were able to increase the levels of acetylation of histone H3 or tubulin. These compounds were also assayed in breast cancer MCF7 cells to test their antiproliferative effect. Three compounds showed a significant reduction of cancer proliferation, representing valuable starting points for further optimization.

Full text of DOI:10.1016/j.bioorg.2020.104462

Journal Pre-proofs
Investigation of the effect of different linker chemotypes on the inhibition of
Histone Deacetylases (HDACs)
Pasquale Linciano, Rosaria Benedetti, Luca Pinzi, Fabiana Russo, Ugo
Chianese, Claudia Sorbi, Lucia Altucci, Giulio Rastelli, Livio Brasili, Silvia
Franchini
PII:
S0045-2068(20)31760-0
DOI:
https://doi.org/10.1016/j.bioorg.2020.104462
YBIOO 104462
Reference:
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
8 August 2020
1 November 2020
3 November 2020
Please cite this article as: P. Linciano, R. Benedetti, L. Pinzi, F. Russo, U. Chianese, C. Sorbi, L. Altucci, G.
Rastelli, L. Brasili, S. Franchini, Investigation of the effect of different linker chemotypes on the inhibition of
Histone Deacetylases (HDACs), Bioorganic Chemistry (2020), doi: https://doi.org/10.1016/j.bioorg.2020.104462
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. 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.
©
2020 Published by Elsevier Inc.

Investigation of the effect of different linker chemotypes on the inhibition of
Histone Deacetylases (HDACs)
1
1
1
1
2
Pasquale Linciano, Rosaria Benedetti, Luca Pinzi, Fabiana Russo, Ugo Chianese, Claudia
1
*
2
1
1
1*
Sorbi, Lucia Altucci, Giulio Rastelli, Livio Brasili, Silvia Franchini,
1Dipartimento di Scienze della Vita, Università degli Studi di Modena e Reggio Emilia, via G.
Campi 103, 41125, Modena, Italy.
2Dipartimento di Medicina di Precisione, Università degli Studi della Campania “Luigi Vanvitelli”,
via L. De Crecchio 7, 80138 Napoli, Italy.
Abstract
Histone Deacetylases (HDACs) are among the most attractive and interesting targets in anticancer
drug discovery. The clinical relevance of HDAC inhibitors (HDACIs) is testified by four FDA-
approved drugs for cancer treatment. However, one of the main drawbacks of these drugs resides in
the lack of selectivity against the different HDAC isoforms, resulting in severe side effects. Thus,
the identification of selective HDACIs represents an exciting challenge for medicinal chemists.
HDACIs are composed of a cap group, a linker region, and a metal-binding group interacting with
the catalytic zinc ion. While the cap group has been extensively investigated, less information is
available about the effect of the linker on isoform selectivity. To this aim, in this work, we explored
novel linker chemotypes to direct isoform selectivity. A small library of 25 hydroxamic acids with
hitherto unexplored linker chemotypes was prepared. In vitro tests demonstrated that, depending on
the linker type, some candidates selectively inhibit HDAC1 over HDAC6 isoform or vice versa.
Docking calculations were performed to rationalize the effect of the novel linker chemotypes on
biologic activity. Moreover, four compounds were able to increase the levels of acetylation of
histone H3 or tubulin. These compounds were also assayed in breast cancer MCF7 cells to test their
1

antiproliferative effect. Three compounds showed a significant reduction of cancer proliferation,
representing valuable starting points for further optimization.
Keywords: HDAC, HDAC inhibitors, linker portion, hydroxamic acids, anticancer drugs.
1
. Introduction
Histone deacetylases (HDACs) and histone acetyl-transferases (HATs) play a pivotal role in
epigenetic modification of gene expression [1]. In particular, HDACs catalyze the deacetylation of
the ε-amino lysine residues on histone tails, causing chromatin condensation and transcriptional
repression [2]. Interestingly, the functions of HDACs are not solely restricted to chromatin
remodelling. HDACs are also involved in the deacetylation of non-histone proteins such as
transcription factors (p53), tubulin, chaperone proteins (Hsp90), and other cytoplasmic proteins
(STAT3, BAX, and HMGB1), resulting in the control of cellular proliferation, protein degradation,
signal transduction, apoptosis and inflammation [3], [4]. At present, 18 HDACs have been
identified and classified into four classes: I (HDACs 1, 2, 3, and 8), II (HDACs 4, 5, 6, 7, 9, and 10)
2
+
and IV (HDAC11), which are Zn -dependent enzymes, whereas class III, better known as sirtuins
+
(
SIRT1-7) are NAD -dependent HDACs [5–7]. The overexpression of HDACs, with the
consequent reduction of histone acetylation and transcriptional dysregulation, has been observed in
several diseases. Thus, HDAC inhibitors (HDACIs) have gained increasing interest as promising
agents for the treatment of cancer, neurodegenerative disorders, inflammation and HIV infections
[8–14]. At present, four HDACIs have been approved by the Food and Drug Administration (FDA)
to treat some neoplastic diseases (Figure 1). Vorinostat (SAHA), a pan-inhibitor against class I, II,
and IV, was the first HDACIs approved in 2006 to treat relapsed and refractory cutaneous T-cell
lymphoma (CTCL) [15]. Romidepsin was licensed in 2009 for the treatment of CTCL and in 2011
for peripheral T-cell lymphoma (PTCL) [16]. Lastly, Belinostat and Panobinostat were approved in
2

2
014 and 2015, respectively, for relapsed multiple myeloma [17]. Other HDACIs are currently in
advanced clinical trials as anticancer agents [18–21].
O
O
H
OH
N
OH
N
H
N
N
H
H
O
HN
Vorinostat (SAHA)
Panobinostat
O
O
O
HN
H
O
O
S
N
OH
O
NH
S
N
NH
S
O
H
H
N
O
O
Belinostat
Romidepsin
Figure 1. HDACIs approved for clinical use.
Although clinically effective HDACIs have been approved, one of the main drawbacks resides in
the lack of selectivity against the different HDAC isoforms. This results in severe side effects,
including thrombocytopenia, anorexia, and cardiotoxicity. Thus, the identification of selective
HDACIs represents a challenge for the medicinal chemists in order to minimize toxicity and off-
target effects [22].
The well-accepted pharmacophoric model for the design of novel HDACIs is composed of: i) a cap
group, interacting with the surface of the enzyme; ii) a linker, fitting the tubular access to the zinc
atom, and iii) a metal-binding group, coordinating the catalytic zinc ion (zinc-binding group, ZBG)
[23,24]. The recent literature shows that isoform selectivity can be achieved in many different ways,
focusing mainly on the cap-group and, to a lesser extent, on the ZBG [25,26]. Since the surface
groove of HDACs is significantly different between the eleven isoforms, a wide range of cap-
groups have been proposed to guide the selectivity of the compounds toward the isoform of interest
3

[27,28]. Conversely, less information is available about the effect of the linker portion on isoform
selectivity. Indeed, modifications of the linker portion are more challenging due to the relatively
higher conservation among HDACs substrate-binding tunnel [29]. Therefore only a few chemotypes
have been explored (i.e., alkylic chain of SAHA, cinnamate benzylic ring). Senger et al. took the
first step in this direction by identifying the thiazole and oxazole rings as novel linker moieties able
to direct the selectivity toward the HDAC6 isoform [30]. Thus, with the aim to investigate new
linker chemotypes, a small library of twenty-five hydroxamic acids was prepared and tested. Four
different and hitherto unexplored classes of linkers were considered: (i) aliphatic heterocyclic
linkers such as 1,3-dioxolane and 1,4-dioxa-8-azaspiro[4.5]decane (1-4); (ii) aliphatic linear linkers
such as emisuccinates (5, 5a-c); (iii) aromatic heterocyclic linkers such as benzodioxane,
benzodioxin, benzofuran, and benzotriazole (6-9); (iv) aromatic/unsaturated linear linkers such as
phenyl-oxyacetates/-thioacetates (10, 10a-b, 11, 11a-b) and alkynes (12, 12a; Figure 2).
4

Figure 2. Chemical structures of the newly synthesized HDAC inhibitors.
5

At first, we focused our work on the search of the minimal structural requirements capable of
showing a first sign of activity and selectivity by preparing compounds exclusively formed by the
novel linker chemotypes and the hydroxamic acid as ZBG. Then, selected linkers were decorated
with the cap groups to validate the novel scaffolds. Twenty-five (1-5, 5a-c, 6-10, 10a-b, 11, 11a-b,
1
2, 12a) hydroxamic acids were synthesized and characterized. All the synthesized compounds
were evaluated for their inhibitory activity against purified HDAC1, and HDAC6 enzymes, chosen
as representatives of HDACs class I and IIb. The most active compounds were then tested in MCF7
cells to measure histone H3 and α-tubulin acetylation. Moreover, docking calculations were
performed to rationalize the observed selectivity profile towards HDAC1 and HDAC6 isoforms.
1
. Results and Discussion
.1. Chemistry
1
The synthesis of compounds with an aliphatic heterocyclic linker, such as the 1,3-dioxolanes (1-3)
and 1,4-dioxa-8-azaspiro[4.5]decane (4, 4a-b), was performed as reported in Scheme 1. The
commercially available n-butyl acrylate was oxidized to the corresponding diol (14) in neutral
conditions using potassium permanganate and TEBAC in dry acetone. The diol (14) was then
cyclized with the appropriate carbonyl compound (benzaldehyde for cis-1 and trans-1; 5-bromo-
picolinaldehyde for cis-2 and trans-2; benzophenone for 3; N-methyl-piperidone for 4; 4-benzoyl-
piperidone for 4a and 4-benzyl-piperidone for 4b), using the Dean-Stark apparatus to trap the
forming water, to provide the corresponding n-butyl ester (15-18 and 18a-b). The latter was finally
converted into the hydroxamic acids 1-4 and 4a-b by S Ac with 50% aq. hydroxylamine, in
N
ethanol, at room temperature.
6

OH
O
O
a
b
HO
O
O
R'
R
O
O
O
O
14
15
R = Ph; R' = H
R = 3-Br-Pyrid-5-yl-; R' = H
R, R' = Ph
1
1
1
1
1
6
7
8
R, R' = -CH CH N(CH )CH CH -
2
2
3
2
2
8a R, R' = -CH CH N(COPh)CH CH -
8b R, R' = -CH CH N(CH Ph)CH CH -
2
2
2
2
2
2
2
2
2
c
O
O
OH
R'
N
H
R
O
1
2
3
4
4
4
R = Ph; R' = H
R = 3-Br-Pyrid-5-yl-; R' = H
R, R' = Ph
R, R' = -CH CH N(CH )CH CH -
2
2
3
2
2
a
b
R, R' = -CH CH N(COPh)CH CH -
2 2 2 2
R, R' = -CH CH N(CH Ph)CH CH -
2
2
2
2
2
Scheme 1. Reagents and conditions: (a) KMnO (1.2 equiv.), TEBAC (1.2 equiv.), Acetone, r.t. (3h) 0
4
°
C (30 min.)  r.t. (1h); (b) carbonyl compound (0.75 equiv.), pTSA (cat.), dry Toluene, Dean-Stark trap,
N , rifl., 6-24 h; (c) 50% aq. NH OH (30 equiv.), EtOH, r.t., 24 h.
2
2
The 1,3-dioxane derivative 4c was prepared according to Scheme 2. Condensation of dimethyl-
malonate with 30% aq. formaldehyde in the presence of NaHCO , at room temperature, overnight,
3
led to the diol (19) in 30% yield. The intermediate 19 was cyclized with 4-benzyl-piperidone to
afford the dimethyl ester (20), which was further decarboxylated to monomethyl ester (21) with
NaCl in DMSO at 180°C. Lastly, the ester 21 was converted into hydroxamic acid 4c under the
conditions previously described.
7

O
O
O
O
O
a
HO
O
b
O
O
O
N
O
OH
O
O
O
O
19
20
c
d
O
O
O
O
O
O
O
N
N
HN OH
21
4c
Scheme 2. Reagents and conditions: (a) 30% aq. formaldehyde (3 equiv.), saturated solution of NaHCO3,
r.t., overnight, 30% yield; (b) 4-benzyl-piperidone (0.75 equiv.), pTSA (cat.), N Dean-Stark trap, 24 h,
2
,
4
1
8% yield; (c) NaCl (1.2 equiv.), DMSO, 180 °C, 4 h, 87% yield; (d) 50% aq. NH OH, EtOH, r.t., 24 h,
2
2% yield.
The synthesis of compounds with the aliphatic linear linker (5, 5a-b) is reported in Scheme 3. The
aliphatic linker was prepared by reaction of succinic anhydride with butanol, in dry toluene, at
9
0°C, overnight, to give the methyl hemisuccinate 22. The free carboxylic group of 22 was reacted
with 1-benzylpiperazine (for 23), N-benzoylpiperazine (for 24), or 4-benzylpiperidine (for 25),
using EDC and HOBt as coupling agents in DMF, to provide the corresponding amides (23-25)
[31]. Unexpectedly, the reaction of 23-25 with 50% aq. hydroxylamine did not lead to the
corresponding hydroxamate due to the parallel attack of the hydroxylamine to the amide carbonyl
group. Nevertheless, using the less reactive O-benzylhydroxylamine in the presence of
trimethylaluminum as an activator of the ester carbonyl group, the protected hydroxamic acid 26-28
were successfully obtained in good yield. The deprotection of the O-benzyl hydroxamates 26-28
was performed by hydrogenolysis over palladium/charcoal, using the ThalesNano H-Cube flow
reactor, to give 5, 5a-b in quantitative yield.
8

X
O
O
a
b
c
O
O
Y
O
O
HO
N
O
O
O
22
23 X = H; Y = N
24 X = O; Y = N
25 X = H; Y = CH
X
c
d
Y
O
HN
O
N
O
N
OH
N
N
H
H
O
O
2
2
2
6 X = H; Y = N
7 X = O; Y = N
8 X = H; Y = CH
5
X
d
Y
O
N
OH
N
H
O
5a X = O; Y = N
5b X = H; Y = CH
Scheme 3. Reagents and conditions (a) ButOH (1.2 equiv.), dry toluene, 90 °C, overnight, 90% yield. (b)
appropriate amine (1 equiv.), EDC∙HCl (1 equiv.), HOBt (1 equiv.), DMF, 0 °C to r.t., overnight, 70%
yield (for 23) and 78% yield (for 24), and 67% yield (for 25). (c) O-benzylhydroxylamine (4 equiv.), 2M
(
CH ) Al in diethyl ether (4 equiv.), dry DCM, r.t., N , 2h, 72% yield (for 26), 39% yield (for 27), and
3
3
2
5
9% yield (for 28). (d) H-Cube ThalesNano H -Pd/C, MeOH, 60 °C, 20 bar, 1 mL / min, quantitative
2
yield.
For the synthesis of 5c, 4-benzoyl-piperidone was first reduced to alcohol 29 using sodium
borohydride in methanol, at room temperature, overnight (Scheme 4). The intermediate 29 was then
converted into the corresponding hemisuccinate 30 by reaction with succinic anhydride in DMF,
under microwave irradiation. The reaction of 30 with O-protected hydroxylamine, followed by
hydrogenolysis, led to the final compound 5c.
9

O
O
O
a
b
c
N
O
N
N
OH
O
O
OH
O
29
30
O
O
d
N
O
N
O
H
N
H
N
O
O
O
OH
O
O
31
5c
Scheme 4. Reagents and conditions: (a) NaBH (1.5 equiv.), methanol, r.t., 12 h, 87% yield; (b) succinic
4
anhydride (2 equiv.), DMF, MW, 150 °C, 150 W, 1.5 h, 57% yield; (c) i.) Ethyl-chloroformate (1.2
equiv.), TEA (1.3 equiv.), dry DCM, N , 0 °C to r.t., 2 h; ii.) O-benzylhydroxylamine (1.2 equiv.), TEA
2
(
1.2 equiv.), dry DCM, r.t., nitrogen atmosphere 2 h, 42% yield; (d) H-Cube, 10% Pd/C cartridge, H , 20
2
psi, 60 °C, ethanol, quantitative yield.
The synthesis of compounds with an aromatic heterocyclic linker, such as benzodioxane (6) and
benzodioxine (7), is described in Scheme 5. The commercially available (2,3-
dihydrobenzo[b][1,4]dioxin-2-yl)methanol was readily oxidized to carboxylic acid 32 with alkaline
potassium permanganate, in water, at 60 °C. The acid group was then activated by reaction with
ethyl chloroformate to generate in situ a mixed anhydride which reacted quickly with
hydroxylamine hydrochloride to give the hydroxamic acid 6. For the synthesis of the benzodioxine
7
, the carboxylic function of 32 was protected first by esterification with thionyl chloride, in ethanol
to provide the corresponding ethyl ester 33.
1
0

O
O
O
O
b
a
OH
O
O
OH
OH
N
H
O
O
32
6
c
O
O
d
O
O
O
O
O
O
33
34
e
O
O
OH
N
H
O
7
Scheme 5. Reagents and conditions: (a) KMnO (1.4 equiv.), KOH (0.5 equiv.), H O, r.t. to 60 °C, 4h,
4
2
+
-
7
3% yield. (b) i. Ethyl chloroformate (1.2 equiv.), DCM dry, N , 0 °C, 1 h; ii. K NHOH (2 equiv.),
2
ethanol, 0 °C, 10 min. (c) SOCl (1 mL), EtOH, r.t., overnight, 92% yield. (d) i. N-Bromosuccinimide (2
2
equiv.), AIBN (0.30 equiv.), CCl , reflux, N , h.v., 12h; ii. NaI (3 equiv.), acetone, reflux, 2h, 51% yield
4
2
over two steps. (e) NaOH (8 equiv.), 50% w/w aqueous hydroxylamine (42 equiv.), THF/EtOH 1:1, 0 °C
to r.t., 10 minutes, 63% yield.
The oxidation of the benzodioxane 33 to benzodioxine 34 was performed using one pot reaction. 33
was first brominated with N-bromo-succinimide (NBS) and catalytic AIBN as a radical initiator,
under UV irradiation, in carbon tetrachloride. Thus the di-halogenated benzodioxane (not isolated)
was de-brominated in situ with NaI, in refluxing acetone, to afford 34 in 51% yield.
The synthesis of the benzofuran derivative 8 was performed as depicted in Scheme 6.
Salicylaldehyde was reacted first with ethyl bromoacetate in standard S 2 conditions followed by
N
spontaneous intramolecular cyclization to give ethyl benzofuran-2-carboxylate, which was directly
hydrolyzed to carboxylic acid 35 by refluxing alkaline water. The activation of the carboxylic acid
1
1

with ethyl chloroformate, followed by the addition of potassium hydroxamide, afforded the desired
hydroxamic acid 8.
O
OH
a
O
b
O
O
O
O
O
OH
O
HN OH
35
8
Scheme 6. Reagents and conditions: (a) i. K CO (2 equiv.) and ethyl 2-bromoacetate (1.1 equiv.), DMF,
2
3
1
50 °C. ii. H O, reflux, 4 h, 60% yield over two steps. (b) i. Ethyl chloroformate (1.2 equiv.), DCM dry,
2
+
-
N , 0 °C, 1 h; ii. K NHOH (2 equiv.), ethanol, 0 °C, 10 min, 31% yield.
2
For the synthesis of compound 9, the commercially available benzotriazole-5-carboxylic acid was
first esterified as previously reported. The following addition of a solution of freshly prepared
sodium hydroxyamide provided the corresponding hydroxamate 9 (Scheme 7).
O
O
O
a
b
OH
N
N
N
OH
O
N
N
N
N
H
N
H
N
H
N
H
9
Scheme 7. Reagents and conditions (a) SOCl (1 mL), EtOH, r.t., overnight. (b) NaOH (8 equiv.) in 50%
2
w/w aqueous hydroxylamine (42 equiv.), EtOH/THF 1:1, 0 °C to r.t, 10 minutes, 35% yield.
The synthesis of compounds with an aromatic linear linker (10 and 11) was easily obtained, as
depicted in Scheme 8. Phenol (for 10) or thiophenol (for 11) were reacted with ethyl bromoacetate
in standard S 2 conditions, using potassium carbonate as base. The experimental conditions were
N
slightly modified according to the reactivity of the substrate. The less reactive phenol was reacted in
DMF at 60 °C to provide the ethyl phenoxyacetate 36 in quantitative yield [32]. On the contrary, the
less stable thiophenol was reacted in acetone, at room temperature, to avoid the formation of the
1
2

disulphide under alkaline conditions [33]. The ethyl 2-thiophenylacetate 37 was obtained in 54%
yield, alongside with the expected disulphide. Lastly, acetates 36 and 37 were converted into the
corresponding hydroxamic acids 10-11 using a large excess of 50% w/w aqueous hydroxylamine, in
ethanol, at room temperature for 2-24 hours.
X
a
b
H
O
N
X
X
OH
O
O
3
3
6 X = -O-
7 X = -S-
10 X = -O-
11 X = -S-
Scheme 8. Reagents and conditions (a) K CO (2.5 equiv.), ethyl 2-bromoacetate (1.2 equiv.), DMF,
2
3
6
0 °C, 3h, 94% yield (for 36). K CO (2.5 equiv.), ethyl 2-bromoacetate (1.2 equiv.), Acetone, r.t.,
2 3
overnight, 56% yield (for 37). (b) 50% w/w aqueous hydroxylamine (30 equiv.), EtOH, r.t., 2 h, 43%
yield (for 10) or 24 h, 87% yield (for 11).
For the synthesis of compounds 10a and 11a, bearing a cap-group, 4-hydroxybenzoic acid or 4-
mercaptobenzoic acid was first reduced to the corresponding benzyl alcohol 38 or 39, which was
further reacted with ethyl bromoacetate in the presence of a base to give the ester 40 or 41 (Scheme
9
) [33]. The benzyl alcohols 40 and 41 were converted into the benzyl chlorides 42 and 43, and then
reacted with benzylpiperidine to provide the intermediates 44 and 45. These were lastly converted
into the hydroxamate 10a and 11a as previously described.
1
3

O
HO
a
b
c
HO
O
HO
X
XH
8 X = O
O
XH
3
3
40 X = O
41 X = S
9 X = S
O
Cl
d
e
X
O
O
X
N
O
4
4
2 X = O
3 X = S
44 X = O
45 X = S
O
X
OH
N
H
N
1
1
0a X = O
1a X = S
Scheme 9. Reagents and conditions: (a) LiAlH (2.2 equiv.), dry THF, 0 °C, nitrogen atmosphere, 6h,
4
8
6% yield (for 38) and 97% yield (for 39). (b) ethyl 2-bromoacetate (1.2 equiv.), K CO (2.5 equiv.),
2 3
acetone, r.t., 6h, 46% yield (for 40) and 70% yield (for 41). (c) SOCl (1.2 equiv.), DMF (drop), dry
2
DCM, 0 °C, 2h, quantitative yield. (d) 4-benzylpiperidine (1.1. equiv.), K CO (1.5 equiv.), DMSO, r.t.
2
3
to 60 °C, 6h, 23% yield (for 44) and 32% yield (for 45). (e) aqueous hydroxylamine (30 equiv.),
EtOH, r.t., overnight, 59% yield (for 10a) and 37% yield (for 11a).
Compounds 10b and 11b were obtained in three steps (Scheme 10). The commercially available 4-
hydroxybenzoic or 4-mercaptobenzoic acids was first reacted with ethyl bromoacetate to obtain the
intermediate 46 or 47, respectively. The latter were then coupled with benzylpiperidine, using
EDC/HOBt coupling reagents [31], to give the intermediates 48 and 49 which were easily converted
into the hydroxamates 10b and 11b, using the same reaction conditions described above.
1
4

O
O
O
X
a
b
O
N
HO
HO
O
X
XH
O
46 X = O
48 X = O
49 X = S
O
47 X = S
O
c
N
H
N
X
OH
O
1
1
0b X = O
1b X = S
Scheme 10. Reagents and conditions: (a) ethyl bromoacetate (1.2 equiv.), K CO (2.5 equiv.), DMF, r.t.,
2
3
4
8h, 83% yield (for 46) and 77% yield (for 47). (b) 4-benzylpiperidine (1 equiv.), EDC∙HCl (1 equiv.),
HOBt (1 equiv.), DMF, 0 °C to r.t., overnight, 92% yield (for 48) and 49% yield (for 49). (c) aqueous
hydroxylamine (30 equiv.), EtOH, r.t., 24h, 65% yield (for 10b) and 47% yield (for 11b).
The synthesis of the two compounds with an unsaturated linear linker, without (12) and with CAP
group (12a), was performed as reported in Scheme 11 and 12, respectively. 12 was readily obtained
from the commercially available methyl 3-phenylpropiolate by reaction with sodium hydroxyamide
in 40% yield.
O
O
a
OH
O
N
H
12
Scheme 11. Reagents and conditions (a) NaOH (8 equiv.), aqueous hydroxylamine (42 equiv.),
MeOH/THF 1:1, 0 °C, 5 minutes.
Finally, 12a was obtained from the ester 50 which was previously prepared by a three-component
reaction involving methyl propiolate, benzylpiperidine and paraformaldehyde, using copper(I)
iodide as catalyst.
1
5

O
O
O
a
b
+
OH
O
O
N
NH
N
N
H
H
50
12a
Scheme 12. Reagents and conditions (a) paraformaldehyde (2 equiv.), copper(I) iodide (0.1 equiv.),
DMSO, r.t., 30 min., 67% yield (b) 50% w/w aqueous hydroxylamine (3 equiv.), EtOH, r.t., 1h, 63% yield.
1
.2.
HDACs Inhibition Studies
All the synthesized compounds were evaluated in vitro for their inhibitory activity against purified
human HDAC1 and HDAC6 enzymes using a fluorescence-based activity assay [34]. Compounds
1
-5 were tested as racemic mixtures. The compounds were screened first at 50 µM, and the results
were expressed as % of the inhibitory activity. The compounds showing an inhibition >50% were
further assessed at lower concentrations (10 and 1 µM). Vorinostat (SAHA), the pan-inhibitor, and
tubastatin, an HDAC6 selective inhibitor, were used as reference compounds at 5 µM
concentration. The inhibitory activity of all compounds is reported in Table 1.
Table 1. Inhibitory activity of the tested compounds against purified
HDAC1 and HDAC6 enzymes.
%
inhibition HDAC1a
%inhibition HDAC6a
Compound
5
0 M 10 M 1 M 50 M 10 M 1 M
Cis-1
Trans-1
Cis-2
Trans-2
3
33±8
-
-
-
-
37±9
44±5
75±2
73±2
31±2
50±7
6±9
11±5
40±2
79±8
74±4
83±7
72±9
20±9
56±2
87±2
80±4
78±4
51±6
-
-
-
-
25±9
63±3
87±7
41±9
70±4
67±8
83±6
91±4
31±3
48±3
74±3
50±1
78±2
90±5
77±1
82±5
82±5
96±8
49±1
56±5
-
47±1
48±2
41±6
57±5
-
-
51±2
-
-
-
55±1
63±3
-
-
-
1±2
3±2
-
-
-
4
4
a
4
b
-
-
-
-
-
4
c
47±5
-
-
5±4
-
5
a
67±5
69±2
75±7
59±1
-
50±2
76±9
68±4
56±5
31±5
62±5
2±4
54±4
33±1
-
17±1
2±1
11±2
3±2
-
5
-
5
b
59±1
25±6
62±3
72±1
62±2
64±4
57±8
87±9
5
c
6
7
8
9
0
51±8
60±1
50±1
38±1
49±6
67±4
1
1
0a
1
6

1
0b
85±5
88±2
92±4
89±9
12±5
40±6
85±2 18±14
87±3
87±4
63±9
87±1
77±2
92±8
29±4
65±6
52±7
59±8
41±6
64±4
95
-
1
1
73±1
92±3
79±9
-
55±1
53±8
51±4
-
3±1
10±1
4±1
-
1
1ab
1
1b
1
2
1
2a
-
-
3±1
SAHAc
Tubastatinc
100
51
89
a
Values were measured in triplicates. Percentages of inhibition are reported as
b
c
mean ± SD. compound 11a, IC50 = 0.58 µM ± 0.13 against HDAC1 Tested at
µM for both HDAC1 and HDAC6.
5
The tested compounds showed an enzymatic inhibitory activity in the micromolar range, with a
percentage of inhibition at 50 µM ranging between 12-88% against HDAC1 and 6-87% against
HDAC6. Despite the moderate inhibitory activity, our first goal was to understand if selectivity
could be achieved by modulating the linker portion. Indeed these results demonstrate that,
depending on the linker type, some candidates show an activity profile shifted toward HDAC1 over
HDAC6 (compounds trans-2, 7, 10a, 11a) or vice versa (5, 5b). Within the series of HDAC1
selective inhibitors, 11a is the most potent with an IC value of 0.58 μM (see Figure SI-1 in the
5
0
Supporting Information).
Therefore, these novel linker chemotypes, adequately decorated with selective cap groups, could
represent valuable starting points for further hit optimization in the search of more potent and
isoform-selective HDACs inhibitors. Docking calculations were performed to rationalize the effect
of the novel linker on biologic activity and selectivity.
1
.3. Molecular modelling
Although HDAC1 and HDAC6 belong to the same protein family, their binding sites present
peculiar features, resulting in different ligand selectivity [28]. Superimposition of the 4BKX
(HDAC1) [35] and 5EDU (HDAC6) [36] crystal structures (see Figure SI-2 in the Supporting
Information) shows that the HDAC1 and HDAC6 active sites present similar residues. However,
1
7

compared to HDAC6, HDAC1 shows a narrower pocket at the cap region entrance but a wider
pocket in the zinc and linker regions, thus resulting in different active site shapes. The latter feature
is due to a broader separation of Phe150 and Phe205 residues in HDAC1 with respect to the
corresponding Phe620 and Phe680 residues of HDAC6. To help rationalize the effect of structurally
different linkers on the selectivity of the synthesized compounds, docking calculations were
performed into the 4BKX [35] and 5EDU [36] crystal structures, as described in the Methods
section. Visual inspection of the predicted docking poses in the two targets shows that the
hydroxamic acid moiety of the investigated compounds coordinates the catalytic Zn2+ ion, in line
with other inhibitors bearing this ZBG [37]. To evaluate the effect of stereochemistry on enzymatic
activity, the diastereomeric couple cis/trans-1 and 2 were prepared and tested. According to the
predicted docking poses, the 1,3-dioxolane moiety of cis-2 and trans-2 binds near to the Phe150,
His178, Phe205, and Tyr303 (HDAC1), and the corresponding Phe620, His651, Phe680, and
Tyr782 residues in HDAC6. Moreover, the pyridine nitrogen of these compounds is located near
Phe205 and Leu749 in HDAC1, and the Phe680 and Leu271 residues in HDAC6, with the bromine
substituent extending toward the outer enzyme surface. Indeed, trans-2 (Figure 3, panels a and b)
fits slightly better than cis-2 within the HDAC1 and HDAC6 binding sites due to the 2R and 4S
stereochemistry of the 1,3-dioxolane ring, which results in lower steric hindrance. Interestingly,
compounds bearing a bulkier linker portion, namely 1,4-dioxa-8-azaspiro (4-4b) and the 1,5-dioxa-
9
-azaspiro (4c), docked better into the HDAC1 active site, mainly because of the wider pocket of
HDAC1 in the zinc and linker binding regions described above. Docking calculations and in vitro
assays also suggested that compounds 5-5c could not bind efficiently to the hydrophobic tunnel of
HDAC1 and reach the catalytic Zn2+ ion, due to the steric clashes of the piperazine or piperidine
rings with the Phe150 and Phe205 side chains. On the contrary, favourable binding modes, of e.g.
compound 5, were found in HDAC6 (Figure 3, panel d), which has a wider entrance in the cap
region. As a result, ligands bearing these linker chemotypes are inactive or significantly less active
against HDAC1 compared to HDAC6 (e.g., 5, 5a, 5b and 5c). Partially unsaturated aromatic
1
8

heterocyclic linkers, such as the 1,4-benzodioxane moiety of compound 6, is well accommodated in
the Phe150, His178 and Phe205 tunnel of HDAC1. This compound could not dock into the
narrower hydrophobic tunnel of HDAC6, in line with the observed selectivity for HDAC1, due to
steric clashes between the 1,4-benzodioxane moiety and the Phe620, His651 and Phe680 side
chains of HDAC6 (Figure 3, panels g and h). Molecules with aromatic heterocyclic linkers directly
attached to the hydroxamate (e.g., 7, 8 and 9) could dock in both HDAC1 and HDAC6 binding
sites. This result is in agreement with their enzymatic inhibitor profile. An example of the binding
mode is shown in Figure 3, panel e and f, for compound 7. Compounds based on
aromatic/unsaturated linkers, such as the 2-phenoxyethane (e.g., 10-10b) and 2-(phenylthio)ethane
(e.g., 11-11b) moieties, resulted to be more active on HDAC1. According to the predicted binding
2
+
modes, the hydroxamic acid of these compounds coordinates the catalytic Zn ion in HDAC1, and
the phenyl ring establishes favorable π-π stacking interactions with the Phe150 and Phe205
residues. Figure 3, panel c, shows the binding mode of 10 into HDAC1. Compound 12 with an
alkyne linker and a phenyl ring could dock better in HDAC6, due to -  interactions with the two
phenylalanine residues.
1
9

2
0

Figure 3. Binding modes evaluated for representative compounds of the series into 4BKX (HDAC1) and
EDU (HDAC6). 4BKX and 5EDU are coloured in pink and deep teal, respectively. Predicted docking poses
5
are represented in dark grey sticks. Panels a, c, e, and g show the predicted binding mode for trans-2, 10, 7,
and 6 into 4BKX, respectively. Panels b, d, f, and h show the predicted binding mode for trans-2, 5, 7, and 6
into 5EDU, respectively. The image was created with PyMOL (The PyMOL Molecular Graphics System,
Version 1.8, Schrödinger, LLC).
2
.4. Cellular studies
Based on in vitro inhibition studies, the selected compounds cis-2, trans-2, 5, 5b, 7, 8, 10a, 10b,
1a, and 11b, showing an activity >50%, were further tested in MCF7 cells to evaluate their effect
1
on the acetylation levels of histone H3, one of the main histones under control of HDAC1, and -
tubulin (HDAC6 target). Compounds that showed a significant increase of the acetylation levels at
5
0 µM were further investigated at the lower concentrations of 10 µM and 1 µM (Figure 4).
2
1

Figure 4. Western blotting analysis of the tested and reference compounds SAHA and Tubastatine on
acetylation levels of histone H3 and α-tubulin in MCF7 cells. GAPDH was used as a loading control.
Densitometric values were analyzed using ImageJ Software and are reported on the top of bands as the ratio
between loading control and acetylated target.
In the series of compounds with an aliphatic heterocyclic linker, trans-2 was the only compound
that slightly increased the acetylation of histone H3 at 50 µM (2.5-fold vs control) whilst the effect
was not retained at lower concentration (10 µM). As expected, the corresponding cis-2 isomer, less
active than the former, showed a reduced level of K14acH3 (1.5-fold). Conversely, regarding the
acetylation of -tubulin, a comparable increase was observed for both cis- and trans-2 isomers.
This is in agreement with their inhibitor activity against HDAC-6 at 50 µM (75% and 73% for cis-2
and trans-2, respectively).
2
2

In the series of compounds with an aliphatic linear linker, 5 and 5b were chosen for their selectivity
profile slightly shifted toward the HDAC-6 isoform (62% and 54% of inhibitor activity at 1 µM,
respectively). Unexpectedly, at the cellular level, they did not show an appreciable increment of -
tubulin acetylation at 50 µM concentration. The lack of activity, in the cellular compartment, could
be due to the low permeability of these compounds across the plasmatic membrane, probably
associated with the high polarity of the piperazinyl moiety. Due to the promising selectivity profile
of this series, further SAR studies are needed to better understand the activity/efficacy profile.
In the series of compounds with an aromatic heterocyclic linker, the two constrained analogues of
1
0, the benzodioxine 7, and the benzofuran 8 were able to promote the expression of K14acH3, 7
being more effective than 8 and showing efficacy in a dose-dependent manner (50, 10 and 1 µM).
In addition, 7 showed a modest increment of the level of acetylated -tubulin at 50 µM. These
results are in accordance with the enzymatic inhibitor profile of 7 that is a valuable inhibitor of
HDAC-1 (60% of inhibition at 1 µM), whereas it is less effective against HDAC-6.
In the series of compounds with an aromatic-linear linker, 10a-b (phenoxy acetate) and 11a
(phenylthio acetate) are noteworthy for their highest efficacy, with a 12.6-15.6-fold increased level
of histone H3 acetylation at 50 µM. This dose-dependent effect persists at the lower concentrations
of 1 µM. Although to a lesser extent, 10a was also able to modulate the expression of acetylated
tubulin at 50 µM (4.35-fold) and 10 µM (2.6-fold). Overall, the efficacy of this series of compounds
correlates well with isoform selectivity, which is mainly shifted towards HDAC1.
Lastly, 7, 10a, and 11a, the best hit compounds in terms of enzymatic activity and cellular efficacy
were selected to assess their antiproliferative activity in human breast cancer cell lines (MCF7).
Cell viability was measured at 24, 48, and 72 h. As displayed in Figure 5, all compounds showed a
significant reduction of cancer proliferation in a time and dose-dependent manner. The
antiproliferative activity exerted by these compounds is in agreement with their inhibitory activity
that is, in part, associated with the modulation of HDAC1 and HDAC6 isoforms.
2
3

Figure 5. Cell viability assay (MTT) of the test compounds in human breast cancer cell lines (MCF7). Cells
were treated with 3 increasing concentrations (1, 10, and 50 M) of tested compounds and monitored at 24,
4
8, and 72 hours. Data are expressed as mean ± SEM for three independent experiments, each performed in
duplicate.
3
. Conclusions
In conclusion, a series of twenty-five novel HDACs inhibitors bearing unexplored linker
chemotypes were synthesized and evaluated for their inhibitory activity against the HDAC1 and
HDAC6 isoforms. Bulky (aliphatic or aromatic heterocyclic) linkers favor HDAC1 selectivity while
aromatic linear (phenoxyacetate/thioacetate) linkers shift the selectivity towards HDAC6. Docking
studies were performed to rationalize the effect of the novel linker chemotypes on inhibitory
activity and selectivity. So far, the effect of different linker chemotypes has been relatively less
explored. Our study shows that, depending on the linker, selectivity can be directed toward HDAC1
or HDAC6 isoform, thus giving the opportunity to medicinal chemists to exit from the conventional
chemical space widely explored for HDACs inhibitors. The best compounds were tested in a cell-
based study to evaluate the acetylation levels of H3 (HDAC1 target) or -tubulin (HDAC6 target).
Increased levels of acetylation were seen for trans-2, 7, 10a, 11a. In addition, compounds 7, 10a,
and 11a showed a significant reduction of proliferation in MCF7 human breast cancer cell line.
2
4

Taken together, these results will set the basis for advancing the design and discovery of more
potent and selective HDACs inhibitors to fight cancer.
4
. Experimental
4
.1. Chemistry
All the reagents, solvents, and other chemicals were used as purchased without further purification
unless otherwise specified. Air- or moisture-sensitive reactions were performed under argon
atmosphere. Reactions were monitored by thin-layer chromatography on silica gel plates (60F-254,
E. Merck) and visualized with UV light, cerium ammonium sulfate, alkaline KMnO4 aqueous
solution, or 1% FeCl ethanolic solution. Column liquid chromatography (LC) purifications were
3
carried out using Merck silica gel 60 (230-400 mesh, ASTM). Flash chromatography purifications
were performed with the ISOLERA-Biotage system. All the hydrogenations were performed with
the ThalesNano H-Cube Mini Plus flow reactor. The structures of all isolated compounds were
1
13
ensured by nuclear magnetic resonance (NMR) and mass spectrometry. H and C NMR (1D and
D experiments) spectra were recorded on a DPX-400 Avance (Bruker) spectrometer at 400 MHz
or on a DPX-600 Avance (Bruker) spectrometer at 600 MHz. Chemical shifts are expressed in ppm
2
(
δ) and calibrated on the residue signal of the solvent: CDCl δ 77.04, CD OD δ 49.8, DMSO-d δ
3 3 6
3
9.5. NMR data are reported as follows: chemical shift, multiplicity (s, singlet; d, doublet; t, triplet;
q, quartet; qnt, quintet; sxt, sextet; m, multiplet; br, broad), coupling constants (Hz) and number of
1
1
1
13
protons/carbons. H- H correlation spectroscopy (COSY), H- C heteronuclear multiple quantum
coherence (HMQC) and heteronuclear multiple bond connectivity (HMBC) experiments were
1
1
1
13
recorded for the determination of H- H and H- C correlations, respectively. Elemental analysis
was performed on a C, H, N, S CE Instruments EA 1110. The following solvent, reactive and
chemical moieties were abbreviated: ethyl acetate (AcOEt), dimethylsulfoxide (DMSO),
dichloromethane (DCM), cyclohexane (CE), diethyl ether (Et O), methanol (MeOH), ethanol
2
2
5

(EtOH), tetrahydrofuran (THF), dimethylformamide (DMF), triethylamine (TEA), 1-Ethyl-3-(3-
dimethylaminopropyl)carbodiimide (EDC), hydroxybenzotriazole (HOBt), N-bromosuccinimide
(
NBS), azobisisobutyronitrile (AIBN), piperidine (Pip), piperazine (Pipz), benzyl (Bn), benzoyl
Bz), furan (Fur), diox (Dioxane), dioxi (Dioxin).
(
4
4
.1.1. General procedure for the synthesis of hydroxamic acids 1-12
.1.1.1. METHOD A: To a solution of aliphatic esters (1equiv.) in ethanol, NH OH 50% w/w in
2
H O (30 equiv.) was added at room temperature. The reaction was stirred for 1-48 hours and
2
concentrated. The solvent was evaporated under reduced pressure, and the residue was diluted with
water. The aqueous solution was neutralized to pH 7 with 1N aqueous HCl and extracted with
AcOEt. The organic layer was dried over anhydrous Na SO and concentrated. The final compound
2
4
was purified by chromatography or crystallization.
.1.1.2. METHOD B: To a solution of carboxylic acid (1 equiv.) in dry DCM, at 0°C and under
4
nitrogen atmosphere, ethyl chloroformate (1.2 equiv.) was added. The mixture was stirred in the
same conditions for 1 hour. Thereafter, an ethanolic solution of potassium hydroxamide, prepared
by dissolving hydroxylamine hydrochloride (2 equiv.) and potassium hydroxide (2 equiv.) in
ethanol, was added dropwise to the first solution at 0 °C. The reaction was stirred for a further 15
minutes and concentrated. The solvent was evaporated under reduced pressure and the residue was
diluted with water. The aqueous solution was neutralized to pH 7 with 1N aqueous HCl and
extracted with AcOEt. The organic layer was dried over anhydrous Na SO and concentrated. The
2
4
final compound was purified by chromatography or crystallization.
.1.1.3. METHOD C: NaOH (8 equiv.) was solubilized in 50% w/w aqueous hydroxylamine (42
4
equiv.) at 0 °C. The appropriate aromatic ester (1 equiv.) solubilized in THF/EtOH 1:1 was added
dropwise, and the mixture stirred at room temperature for 10 minutes. The reaction was quenched
with glacial acetic acid (8 equiv.) and concentrated under reduced pressure. The residue was diluted
with water and neutralized with 1N aqueous HCl. The aqueous solution was extracted in AcOEt,
2
6

and the organic phase washed with NH Cl, dried over anhydrous Na SO and concentrated. The
4
2
4,
final compound was purified by chromatography or crystallization.
4
.1.1.4. METHOD D: Appropriate O-benzylhydroxylamine derivates were solubilized in MeOH,
and the hydrogenation was performed with H-Cube Mini Plus ThalesNano with the following
conditions: Temperature 60 °C, H 20 psi, cartridge Pd/C, solvent MeOH, flow 1mL/min. The
2
solvent was concentrated to give pure hydroxamic acid.
4
.1.1.1.1. Cis-N-hydroxy-2-phenyl-1,3-dioxolane-4-carboxamide (cis-1)
According to Method A. Chromatographed over silica gel: ratio crude/silica gel 1:100 DCM 100%.
1
TLC: DCM:MeOH 95:5 Rf=0.34. Pale yellow liquid (66%yield). H NMR (400MHz, DMSO-d )
6
HD48_C δ 4.25 (t,1H, J = 2.0 Hz, O-CH CH); 4.50 (dd,1H, J = 2.0 Hz, OCH CH); 4.71 (dd,1H, J =
2
2
1
3
1
.0 Hz, OCHCH ); 5.83 (s,1H, OCHO); 7.50-7.43 (m, 5H, CH ); 8.83 (s,1H, NH). C NMR
2 Ar
(
100MHz, DMSO-d ) δ 69.46 (O-CH -CH); 74.14(CH -CH-O); 105.46(O-CH-O); 126.51 (CH );
6 2 2 Ar
1
5
.28.80(CH ); 130.09(CH ); 135.61(C ); 168.15 (CO). Anal. Calcd. for C H NO : C, 57.41; H,
Ar Ar Ar 10 11 4
.30; N, 6.70; Found C, 57.40; H, 5.45; N, 6.72.
4
.1.1.1.2. Trans-N-hydroxy-2-phenyl-1,3-dioxolane-4-carboxylate (trans-1)
According to Method A. Chromatographed over silica gel: ratio crude/silica gel 1:60 DCM/MeOH
1
9
5:5. TLC: DCM:MeOH 95:5 Rf=0.28. Orange liquid (56% yield). H NMR (400MHz, DMSO-d )
6
HD49_A δ 3.95 (dd, 1H, J = 6.1 Hz, 8.3 Hz, OCH CH); 4.35 (t, 1H, J = 8.0 Hz, OCH CH); 4.60 (t,
2
2
1
9
1
H, J = 8.0 Hz, OCHCH ); 5.91 (s, 1H, OCHO); 7.45 – 7.35 (m, 3H); 7.54 – 7.44 (m, 2H), 9.45-
2
1
3
.96 (brs, 2H, NH, OH). C NMR (100MHz, DMSO-d ) δ 67.97 (OCH CH); 73.51 (CH CHO);
6
2
2
03.89 (OCHO); 126.67 (CH ); 128.17 (CH ); 129.31 (CH ); 137.08 (C );166.46 (CO). Anal.
Ar
Ar
Ar
Ar
Calcd. for C H NO : C, 57.41; H, 5.30; N, 6.70; Found C, 57.43; H, 5.27; N, 6.68.
1
0
11
4
4
.1.1.1.3 Cis- 2-(5-bromopyridin-2-yl)-N-hydroxy-1,3-dioxolane-4-carboxamide (cis-2)
2
7

According to Method A. Chromatographed over silica gel: ratio crude/silica gel 1:80 DCM/MeOH
1
9
5:5. TLC: DCM:MeOH 95:5 Rf=0.30. Yellow solid(49%yield). H NMR (400 MHz, CD OD)
3
HD63_A δ 4.28(dd, 1H, J = 8.0, 12 Hz, OCH CH); 4.37 (d, 1H, J = 8.0 Hz, OCH CH); 4.73 (dd,
2
2
1
8
H, J = 4.0, 8.0 Hz, OCHCH ); 5.90 (s, 1H, OCHO); 7.63 (m, 1H, CH ); 8.13 (m, 1H, CH );
2 Ar Ar
.76(d, 1H, J = 4.0 Hz, CHAr). 13C NMR (100 MHz, CD OD) δ 71.11 (OCH CH); 76.43
3
2
(
OCHCH ); 105.31 (OCHO); 122.97 (BrC ); 125.65 (CH ); 141.61 (CH ); 151.46 (CH );
2 Ar Ar Ar Ar
1
3
55.20 (C ); 170.28(C=O). Anal. Calcd. for C H BrN O : C, 37.39; H, 3.14; N, 9.69; Found C,
Ar 9 9 2 4
7.35; H, 3.10; N, 9.61.
4
.1.1.1.4. Trans- 2-(5-bromopyridin-2-yl)-N-hydroxy-1,3-dioxolane-4-carboxamide (trans-2)
According to Method A. Chromatographed over silica gel: ratio crude/silica gel 1:80 DCM/MeOH
1
9
5:5. TLC DCM:MeOH 95:5 Rf=0.27. Yellow liquid (70%yield). H NMR (400 MHz, CD OD)
3
HD64_A δ 4.12 (dd, 1H, J = 8.0, 12 Hz, OCH CH); 4.46 (t,1H, J = 8.0Hz, OCH CH); 4.78 (t, 1H, J
2
2
=
8.0Hz, OCHCH ); 5.99 (s, 1H, OCHO); 7.58 (d, 1H, J = 8.0Hz, CH ); 8.09 (t, 1H, J = 8.0Hz,
2 Ar
1
3
CH ); 8.69(d, 1H, J = 4.0Hz, CH ). C NMR (100 MHz, CD OD) δ 69.90 (OCH CH); 75.84
Ar
Ar
3
2
(
OCHCH ); 105.17 (OCHO); 122.59 (BrC ); 123.99 (CH ); 141.50 (CH ); 151.18 (CH );
2 Ar Ar Ar Ar
1
3
56.18 (C ); 169.50 (C=O). Anal. Calcd. for C H BrN O : C, 37.39; H, 3.14; N, 9.69; Found C,
Ar 9 9 2 4
7.31; H, 3.18; N, 9.65.
4
.1.1.1.5. N-hydroxy-2,2-diphenyl-1,3-dioxolane-4-carboxamide (3)
According to Method A. Chromatographed over silica gel: ratio crude/silica gel 1:60 DCM/MeOH
1
9
5:5. TLC: DCM:MeOH 95:5 Rf=0.35. Beige solid (57% yield). H NMR (400MHz, DMSO)
HD52_A δ 4.04(dd, 1H, J = 8.0, 6.4 Hz, OCH CH); 4.12(t,1H, J = 6.4 Hz, OCH CH); 4.74(t, 1H, J
2
2
1
3
=
6.4 Hz, OCHCH ); 7.31-7.51(m, 10H, CH ); 9.14(brs, 1H, NH); 10.87(brs, 1H, OH). C NMR
2 Ar
(
100 MHz, CDCl3) δ 61.74 (OCH CH); 73.60 (OCHCH ); 110.41 (OCO); 125.73(C );
2 2 Ar
2
8

1
26.11(C ); 127.91(C ); 128.24 (C ); 141.49(C ); 141.70(C ); 165.28(C=O). Anal. Calcd. for
Ar Ar Ar Ar Ar
C H NO : C, 67.36; H, 5.30; N, 4.91; Found C, 57.30; H, 5.35; N, 4.85.
1
6
15
4
4
.1.1.1.6. N-hydroxy-8-methyl-1,4-dioxa-8-azaspiro[4.5]decane-2-carboxamide (4)
According to Method A. Chromatographed over silica gel: ratio crude/silica gel 1:20
1
Acetone/MeOH 6:4. TLC DCM:MeOH 85:5 Rf=0.2. Colorless liquid (20%yield). H NMR (400
MHz, CD OD) HD59_A δ 1.78-2.00 (m, 4H, CH CHCH ) 2.34 (s, 3H, CH N); 2.48-2.80 (m, 4H,
3
2
2 ;
3
CH NCH ); 4.09 (dd, 1H, J = 8.4, 5.6 Hz OCH CH); 4.27 (dd, 1H, J = 8.8, 7.2 Hz, OCH CH); 4.57
2
2
2
2
1
3
(
dd, 1H, J = 7.6, 5.6 Hz, OCHCH ). C NMR (100 MHz, CD OD) δ 34.70 (CH CCH ); 35.50
2 3 2 2
(
CH CCH ); 45.65 (CH N); 54.12 (CH NCH ); 54.22 (CH NCH ); 68.16 (OCH CH); 75.21
2 2 3 2 2 2 2 2
(
OCH CH); 110.02 (CH CCH ); 170.06 (C=O). Anal. Calcd. for C H N O : C, 49.99; H, 7.46; N,
2 2 2 9 16 2 4
1
2.96; Found C, 50.10; H, 7.38; N, 12.90.
4
.1.1.1.7. 8-benzoyl-N-hydroxy-1,4-dioxa-8-azaspiro[4.5]decane-2-carboxamide (4a)
According to Method A. Chromatographed over silica gel: ratio crude/silica gel 1:100 DCM/MeOH
1
9
5:5. TLC DCM:MeOH 95:5 Rf=0.5. Viscous orange liquid (60%yield). H NMR (400 MHz,
CDCl3) HD54_A δ 1.60-2.01 (m, 4H, CH CCH ); 3.35–3.80 (m, 3H, CH NCH ); 3.35-4.15(m, 1H,
2
2
2
2
CH NCH ); 4.17-4.20 (m, 1H, OCH CH); 4.25-4.32 (m, 1H, OCH CH); 4.64 (m, 1H, OCHCH );
2
2
2
2
2
1
3
7
.42 (m, 5H, CH ). C NMR (100 MHz, CDCl3) δ 34.91 (2C, CH CCH from HMQC); 44.45
Ar 2 2
(2C, CH NCH from HMQC); 67.27 (OCH CH); 74.05 (OCH CH); 109.86 (OCCH ); 126.86
2 2 2 2 2
(
CH ); 128.59 (CH ); 129.95 (CH ); 135.48 (C ); 167.93 (NC=0); 173.64 (NHC=O). Anal.
Ar
Ar
Ar
Ar
Calcd. for C H N O : C, 58.82; H, 5.92 N, 9.15; Found C, 58.75; H, 5.98; N, 9.22.
1
5
18
2
5
4
.1.1.1.8. 8-benzyl-N-hydroxy-1,4-dioxa-8-azaspiro[4.5]decane-2-carboxamide (4b)
According to Method A. Chromatographed over silica gel: ratio crude/silica gel 1:80 DCM:MeOH
1
9
:1. TLC DCM:MeOH 95:5; Rf=0.45. White solid (55%yield). H NMR (400MHz, DMSO)
2
9