Synthesis and structure activity relationship of 1, 3-benzo-thiazine-2-thiones as selective HDAC8 inhibitors

Article abstract of DOI:10.1016/j.ejmech.2019.111756

Human histone deacetylase 8 (HDAC8) is a highly promising target for neuroblastoma and other types of cancer. Several HDAC inhibitors are approved for the treatment of special cancer subtypes or are evaluated in clinical trials. By far the most drugs or d

Full text of DOI:10.1016/j.ejmech.2019.111756

Journal Pre-proof
Synthesis and structure activity relationship of 1, 3-benzo-thiazine-2-thiones as
selective HDAC8 inhibitors
Benjamin Wolff, Niklas Jänsch, Wisely Oki Sugiarto, Stefan Frühschulz, Maraike
Lang, Rabia Altintas, Ina Oehme, Franz-Josef Meyer-Almes
PII:
S0223-5234(19)30908-0
DOI:
https://doi.org/10.1016/j.ejmech.2019.111756
EJMECH 111756
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 10 July 2019
Revised Date: 29 September 2019
Accepted Date: 2 October 2019
Please cite this article as: B. Wolff, N. Jänsch, W.O. Sugiarto, S. Frühschulz, M. Lang, R. Altintas,
I. Oehme, F.-J. Meyer-Almes, Synthesis and structure activity relationship of 1, 3-benzo-thiazine-2-
thiones as selective HDAC8 inhibitors, European Journal of Medicinal Chemistry (2019), doi: https://
doi.org/10.1016/j.ejmech.2019.111756.
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.
©
2019 Published by Elsevier Masson SAS.

Synthesis and structure activity relationship
of 1, 3-benzo-thiazine-2-thiones as selective
HDAC8 inhibitors
1
1
1
1
1
Benjamin Wolff , Niklas Jänsch , Wisely Oki Sugiarto , Stefan Frühschulz , Maraike Lang ,
2
,3,4,5
2,3,4
1,*
Rabia Altintas
, Ina Oehme , Franz-Josef Meyer-Almes
1
Department of Chemical Engineering and Biotechnology, University of Applied Sciences,
Haardtring 100, 64295 Darmstadt, Germany.
2
3
Preclinical Program, Hopp Children's Cancer Center at NCT Heidelberg (KiTZ), Germany
Clinical Cooperation Unit Pediatric Oncology, German Cancer Research Center (DKFZ), INF 280, D-
9120 Heidelberg, Germany
6
4
German Cancer Research Consortium (DKTK), Germany
5
Medical Faculty of Heidelberg University, Germany
*
Corresponding author:
Franz-Josef Meyer-Almes
Email: franz-josef.meyer-almes@h-da.de
Phone: +49 61511638406
Fax: +49 61511638404
Keywords:
Neuroblastoma, cancer, drug design, non-chelating inhibitor, drug discovery, histone
deacetylase 8

Abstract:
Human histone deacetylase 8 (HDAC8) is a highly promising target for neuroblastoma and other types
of cancer. Several HDAC inhibitors are approved for the treatment of special cancer subtypes or are
evaluated in clinical trials. By far the most drugs or drug candidates contain a hydroxamate group that
chelates the catalytic zinc ion within HDACs. Most hydroxamate inhibitors are more or less
unselective, although there are considerable exceptions demonstrating the general feasibility to
develop at least HDAC isoenzyme selective inhibitors. In addition, hydroxamates have recently come
under discussion regarding their potential for mutagenicity. Recently, PD-404,182 was discovered as a
selective and potent non-hydroxamate inhibitor of HDAC8. However, this active compound turned out
to be decomposed in the presence of glutathion (GSH).
Here, we describe the synthesis of significantly improved analogs of PD-404,182 that demonstrate
both, great selectivity for HDAC8 and also chemical stability in the presence of GSH. The compounds
are characterized with respect to structure-activity relationship, binding mode and target engagement
in neuroblastoma cells by combining biochemical and biophysical methods with chemoinformatics.

1
. INTRODUCTION
Histone deacetylases (HDACs) are established and validated cancer targets [1-8]. There are 11 human
members of the HDAC super family, which can be functionally divided into the classic HDACs with a
+
catalytic zinc ion at the bottom of the highly conserved binding site and sirtuins that are NAD
dependent. By means of multiple sequence alignment, the classic HDACs are subdivided into class I,
containing HDACs 1,2,3 and 8, class IIa including HDAC4, 5, 7 and 9, class IIb, which consists of
HDAC6 and HDAC10 and class IV with only one member, HDAC11. The sirtuins form class III.
HDAC8 is least related to the other class I HDACs. In fact, its functionality in terms of substrate and
inhibitor recognition resembles that of class II isoenzymes, so that it can be classified at the interface
between class I and class II HDACs. HDAC8 is also special regarding protein-protein and substrate
recognition [9]: Unlike all other HDAC isoenzymes, HDAC8 is not member of a multi-enzyme
complex. Furthermore, only a handful of interaction partners and substrates have been unambiguously
identified. [10-15] In fact, although histone H3- and H4-derived peptides can be deacetylated in vitro,
there is no clear evidence that histones are deacetylated by HDAC8 in vivo [16]. Intensive efforts to
elucidate the physiological role of HDAC8 revealed a pivotal role in several types of cancer [8, 17-
1
9]. The most significant results were reported for the participation of HDAC8 in the development and
maintenance of childhood neuroblastoma [20, 21]. A combination of retinoic acid and a selective
HDAC8 inhibitor showed a synergistic and specific effect on the reduction of tumor cells in BE(2)-C
xenografted NMRi nude mice [20]. The authors could also demonstrate less side effects with a
HDAC8 selective inhibitor compared with vorinostat, a pan-inhibitor that hits various HDAC
isoenzymes. Therefore, there is an urgent need for the development of selective HDAC8 inhibitors.
Most described HDAC8 selective inhibitors contain a hydroxamate group to increase affinity through
chelating the catalytic zinc ion [21]. However, aside the potential to bind to other free or enzyme-
bound zinc ions, hydroxamic acids have raised some concerns about their potential mutagenicity,
particular for the treatment of chronic diseases [22]. To avoid hydroxamate groups, several other zinc
binding groups (ZBG’s) have been explored [23, 24]. One of the most promising potent and selective
HDAC8 inhibitors lacking the hydroxamate moiety are benzothiazine-imines, which have been
recently discovered in our laboratory [25]. Showing extraordinary high potency in the single-digit
nanomolar range in an in vitro biochemical enzyme activity assay, the compounds were about 1000-
fold less active in tumor cell lines. Okazaki et al. showed that the basic benzothiazine-imine (PD-
4
04,182) was converted into the corresponding thiophenol in the presence of GSH [26]. Therefore, it
appears plausible that PD-404,182 is captured by intracellular GSH and only a small portion of the
compound reaches its cellular target. Most recently, we elucidated the mechanism by which PD-
4
04,182 inhibits HDAC8 in great detail demonstrating that this compound is indeed a covalent
modifier of cysteine-residues in HDAC8 [27]. Since the pharmacophore is unprecedented regarding its
unusual chemical structure as well as in vitro potency and selectivity, we aimed at the development of
GSH-resistant analogs of PD-404,182.

In this study, we synthesized two series of 1, 3-benzo-thiazine-2-thione (BTZT) analogs of PD-404,182.
The compounds are characterized regarding their chemical stability in the presence of GSH, HDAC8
selectivity as well as target engagement in neuroblastoma tumor cells. The results are complemented
by structure activity relationship analysis and molecular docking.
2
. RESULTS AND DISCUSSION
.1. Synthesis
2
The synthetic route towards the pyrimido[1,2-c][1,3]benzothiazin-6-imines (n=1) and imidazo[1,2-
c][1,3]benzothiazin-6-imines (n=0) is described in Scheme 1 and essentially follows the procedure of
Mizuhara et al. [28].
Scheme 1: Synthetic route to the 1,3-benzo-thiazine-2-thiones
2
.1.1 General synthesis of the 2-aryl-4,5-dihydro-1H-imidazole (2a–v, n=0) and 2-aryl-1,4,5,6-
tetrahydropyrimidine (3a–u, n=1)
To a solution of the aldehyde 1a–v (1 eq.) in tert-butanol (9.0 ml/mmol) the diamine (1.1 eq.) was
added and the solution was stirred at 70 °C for 30 min. K CO (4 eq.) and I (1.25 eq.) was added at
2
3
2
7
0 °C and the mixture was stirred at this temperature for further 3 h. The mixture was cooled down to
rt and Na S O was added until the iodine color almost disappear. The organic layer was separated and
2
2
3
the solvent was removed in vacuo. The received solid was dissolved in water (7.5 ml/mmol) and
N NaOH was added until pH = 12 – 14. The aqueous layer was separated with CHCl3
2
aq
(
3 × 3.75 ml/mmol), the combined organic layers were dried (Na SO ) and the solvent was removed in
2 4
vacuo. The product can be used without further purification.
2
.1.2 General procedure of the cyclization using carbon disulfide
To a mixture of 2a–v or 3a–u (1 eq.) and NaH (2 eq.) in DMF (3.3 ml/mmol) was added CS (2 eq.)
2
under nitrogen atmosphere. After stirring at 80 °C for 16 h the mixture was concentrated in vacuo. The
product was purified via chromatography (SiO , hexane:ethyl acetate = 4:1, v/v).
2

Detailed synthesis procedures and chemical analytics including NMR-spectra and MS data are
provided in the Supporting Information.
2
.2. Thiol-reactivity of BTZT derivatives
In a previous study we identified the benzothiazine-imine scaffold as a novel pharmacophore for
potent and selective HDAC8 inhibitors [25]. However, compounds from this series were 1000-fold
less potent in cells compared with the biochemical assay. This loss in potency was attributed to thiol-
reactivity and effective degradation in the presence of e.g. GSH. More recently, we elucidated the
details of the reaction mechanism between benzothiazine-imine PD 404,182 and GSH and showed
that this compound is a rather unselective cysteine modifier yielding thiocyanates or mixed disulfides
with the thiophenol metabolite of PD 404,182 in protein context [27]. The compound is chemically
stable in the absence of reducing agents, but is quickly decomposed in the presence of GSH. Despite
the unselective nature of this thiol reactivity, inhibitors of the benzothiazine-imine series showed very
high isoenzyme selectivity for HDAC8. In this study we also provided arguments that the covalent
modification of C153 within the active site pocket is preceded by non-covalent recognition of the
benzothiazine motif, which is supposed to be a key element for selectivity.
Fig. 1: Chemical stability of BTZT analogs (5 mM) with exchanged heteroatoms in the presence of
2
1
eq GSH (10 mM). For comparison purposes: GSH concentrations in animal cells range from 0.5 to
0 mM. The dashed line indicates that the decomposition of PD-404,182 is complete within 24 hours
and occurs much faster than that of the other analogs.
Based on this hypothesis we aimed at shifting the balance between unselective thiol-reactivity and
selective non-covalent molecular recognition towards lower reactivity while preserving isoenzyme-
selectivity for HDAC8. At first we determined the time dependence of the decomposition of PD 404,

1
82 analogs, where the imine nitrogen was exchanged by oxygen (6) or sulfur (5l), or where the sulfur
atom in the thiazine ring was changed to oxygen (7) (Fig. 1). In accordance with our previous study
27], PD 404, 182 was completely decomposed after 24 h in the presence of 10 mM GSH at 20 � (Fig.
). The chemical stability of 6 and 7 is significantly higher with 62 and 48 % of the initial substance
[
1
remaining after 24 h, respectively (Table 1). Compared with these derivatives, the BTZT 5l shows
extraordinary high chemical resistance against GSH with essentially unaffected compound even after
4
8 h incubation time under the same conditions as previously (Fig. 1). Based on these results, we
synthesized 42 further compounds and determined their chemical stability in the presence of 10 mM
GSH concentrations that mimic intracellular conditions. All tested BTZT derivatives are chemically
stable under reducing conditions in the presence of GSH (Table 2).
Table 1: Impact of heteroatom exchange on potency and GSH stability.
IC ( µM)
5
0
GSH-
stability (%)
a
#
HDAC1
3.6 ± 0.8
HDAC4
8.6 ± 1.5
HDAC6
6.7 ± 0.8
HDAC8
b
b
b
0.011 ±
PD-404,182
b
< 5
> 95
62
0.001
b
b
b
b
5
l
> 50
> 50
> 50
2.8 ± 0.3
6
> 50
5.9 ± 0.7
13 ± 1.6
> 50
2.0 ± 0.2
9.5 ± 1.1
7
48
a
Percent parent compound remaining uncomposed after 24 h incubation in the presence of 10 mM
b
GSH, [25].
2
.3. Structure-activity relationship analysis
To evaluate the inhibitory potency and isoenzyme selectivity of BTZT analogs of PD 404,182 for
HDAC8, we measured the activity profile of all BTZT analogs against a panel of HDAC isoenzymes
(
Fig. 2A, Table 2). One of the best BTZT’s, 5b, achieves almost the potency of the standard HDAC8
inhibitor PCI-34051 and is more potent against this enzyme than Trichostatin A (TSA) or
suberoylanilide hydroxamic acid (SAHA) (Fig. 2A). Overall, two series of compounds,

dihydroimidazoles (DHIs, n = 0) and tetrahydropyrimidines (THPs, n = 1), with different heteroring
sizes were synthesized. The BTZT pharmacophore in both series was modified at all aromatic
positions to deduce a structure-activity relationship (SAR). In addition, THP analogs with
dimethylated nitrogen heterocycle and oxygen atoms in place of sulfur atoms were synthesized. Both,
DHIs and THPs, show different activities against HDAC8. 5 THPs show IC -values less than 1 µM,
5
0
while only 2 DHIs are below this threshold and the four most potent compounds (5a, 5b, 5c, 5d) are
from the THP series. This could indicate a slight preference of HDAC8 for the 6-membered
heterocycle.
Fig. 2: Dose-response curves of 5b and reference inhibitors suberoylanilide hydroxamic acid (SAHA),
PCI-34051, Trichostatin A (TSA) against HDAC8.
A methyl- and chlorine scan at the aromatic ring reveals a clear up to 35-fold improvement of
inhibitory activity at the R1- and R2-position upon substitution of hydrogen by methyl or chlorine for
both, the THP- and the DHI-series of the BTZT pharmacophore . The effect of the methyl-substitution
in R2 position is stronger than that of chlorine at the same position for both heterocycle sizes (n=0 and
n=1). The effect of methyl-substitution at R3 is significantly lower, but still beneficial for activity. The
effect of chlorine substitution at this position is different for n=0 and n=1. Methyl-substitutions at ring
position R4 is detrimental for both heterocycle ring sizes, while a chlorine atom at this position shows
an increase in activity for the 5-membered and a decrease for the 6-membered heterocycle.
To determine whether the increase in activity would be due to halogen bonding, positions R1 and R2
were also substituted by the fluorine, bromine and iodine. For the DHI series (n=0) the potency of the
inhibitors increases from F << Cl < Br at the R1 position. For iodine the potency decreases again, but
is still much higher than for the fluorine analog. For the THP-series (n=1) there is a similar but less
pronounced dependency on halogen atoms. The activity increases from fluorine to iodine (F << Cl <
I), while the activity of the bromine THP lies between that of the corresponding fluorinated and the
chlorinated analog. Together with the gain in activity due to a methyl group at position R1, these
findings suggest hydrophobic interactions with a weakly hydrated hydrophobic part of the receptor

surface. The slightly reduced or similar effect of iodine with respect to chlorine could also be due to
sterical reasons so that a halogen bond at the R1 position cannot be fully excluded at this point.

Table 2: Activity profile of BTZTs against a panel of HDAC isoenzymes and chemical stability (in the presence of GSH):
IC (µM)
GSH stability
(%)**
50
#
n*
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
R1
H
R2
H
R3
R4
H
H
H
H
H
H
H
H
H
H
H
H
Cl
F
HDAC1
17.7
HDAC2
>50
HDAC3
>50
HDAC4
HDAC5
HDAC6
HDAC7 HDAC8
4
a
CN
6.8
16.1
0.3
6.2
4.5
-
0.17
0.26
0.42
0.60
1.0
1.0
1.2
1.2
1.3
1.6
1.7
2.3
3.2
6.7
7.0
7.8
-
4
b
H
Br
H
H
>50
38
>50
45
>50
>50
17
5.1
14
>50
20
>50
4.1
>50
11
>50
20
4
>50
14
>95
-
-
>95
>95
>95
4
c
Br
Cl
H
H
>50
>50
>50
>50
>50
>50
>50
>50
9.5
>50
21
>50
>50
>50
4
d
H
H
>50
>50
>50
>50
50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
50
>50
>50
>50
>50
>50
>50
50
>50
>50
>50
>50
50
>50
>50
>50
43
>50
>50
>50
>50
1.7
-
4
e
H
Br
>50
-
>50
-
>50
-
>50
10
-
4
f
H
I
H
4
g
H
F
H
4
h
F
H
Br
4
i
H
CH3
H
H
>50
9
>95
>95
>95
-
>95
>95
-
4
j
H
Cl
4
k
CH3
I
H
H
>50
>50
>50
>50
>50
>50
4
l
H
H
4
m
H
H
H
H
>50
>50
50
16
>50
32
4
n
H
H
>50
>50
-
4
o
H
H
N(CH3)2
H
H
H
4
p
H
OCH3
>50
>50
16
>95

4
q
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
H
H
F
H
Cl
H
OCH3
H
H
>50
>50
>50
>50
>50
>50
>50
12
>50
>50
>50
>50
>50
>50
>50
8.1
>50
>50
>50
>50
>50
>50
>50
20
>50
>50
>50
>50
19
>50
1.9
2.2
>50
>50
>50
>50
>50
>50
20
8
6.8
39
2.7
>50
>50
>50
>50
>50
>50
20
24
6.9
>50
1.8
>50
>50
>50
>50
12
>50
4.7
>50
>50
>50
4.7
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
12
13
13
13
21
>95
>95
>95
>95
>95
>95
-
4
r
s
H
H
4
H
4
t
H
H
H
H
I
H
CH3
H
H
4
u
H
H
4
5
v
a
H
H
CH3
H
31
CH3
H
H
0.057
0.076
0.13
0.13
0.37
0.54
0.83
1.0
1.0
1.1
1.3
1.5
1.6
1.8
2.3
2.7
0.52
5.4
21
30
>50
5
b
H
H
-
5
c
Cl
CH3
H
H
H
H
H
H
H
H
H
H
F
H
H
H
24
14
26
>50
50
1.4
>50
3.8
>95
-
-
5
d
H
H
H
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
3.4
3.2
5
e
H
CN
H
H
5
f
Cl
CF3
H
H
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
29
18
>95
>95
>95
-
>95
>95
>95
>95
>95
>95
-
>95
>95
>95
>95
>95
5
g
H
H
>50
>50
8.1
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
>50
34
>50
>50
>50
>50
>50
>50
15
>50
>50
>50
>50
-
5
h
CH3
H
H
5
i
I
H
5
j
H
Br
F
H
>50
>50
>50
>50
>50
>50
>50
>50
21
5
k
H
H
5
l
H
H
H
5
m
Br
F
H
H
5
n
H
H
5
o
p
q
H
H
H
5
H
Br
H
H
H
H
H
N(CH3)2
H
H
5
H
H
5
r
s
H
Cl
H
H
5
H
F
>50
8.8
>50
>50
>50
>50
>50
-
-
5
t
H
H
Cl
CH3
5
u
H
H
*
*
n = 0: DHI; n = 1: THP
* Percent parent compound remaining uncomposed after 24 h incubation in the presence of 10 mM GSH.

At the R2-position the strongest effects for both, the DHI- and the THP-series, come from methyl-
substitutions and there is no significant gain in activity from fluorine to the heavier halogens.
Therefore, a halogen bond at this position is unlikely. However, a DHI with a bromine substitution at
R2-position shows similar potency than a methyl substitution pointing towards hydrophobic
interactions with HDAC8. For THPs halogen substitution at R2-position does not lead to a significant
improvement. Chlorine- and bromine substituents at the R3-position increase the inhibitory activity
against HDAC8 only for DHIs but not for THPs. The reason behind this divergent behavior could be
due to slightly deviating binding poses. Although, chlorine at R3-positions in THPs seems to have an
adverse effect on activity, the analog with bromine at this positions shows essentially the same activity
than the unsubstituted core compound 5l. Therefore, we reasoned that the halogen substituents in R3
position seem not to be involved in substantial clashes with the receptor surface and that this position
could be further explored with larger substituents for further optimization. Fluorine and chlorine
atoms, but not a methyl group at R4 lead to an increase in potency within the DHI-series.
However, the absolute activities of all DHI-analogs are still less than those for the unsubstituted THP
5
l. Within the THP-series at position R4 the substitution of hydrogen by larger moieties, whether it be
a methyl group or a halogen, causes a pronounced more than 10-fold drop in potency.
In a further attempt we explored the impact of larger cyano- dimethylamino- and methoxy-groups at
position R3 on potency. Interestingly, the cyano-group leads to a considerable increase of activity in
the nanomolar range. In contrast, dimethylamino- and methoxy-substitution in R3 position turned out
to be detrimental. Dimethylation at the R5-position in the 6-membered nitrogen heterocycle of THPs
also led to a decline in activity. Compared with the original substance PD 404,182, the best BTZT 5a
has 5-6 times less activity against HDAC8, but shows dramatically enhanced chemical stability against
GSH, while the selectivity for HDAC8 is preserved (Table 2). Substitution of the sulfur atom within
the thiazine ring of 5l by oxygen causes a 3-4 fold drop in activity and significant decrease in GSH-
stability (Fig. 1, Table 1). The exchange of the thione-sulfur against oxygen leads to similar activities
against HDAC8, but concurrently causes clearly less GSH-stability (Fig. 1, Table 1).
The impact of structural variations on the inhibitory activity of BTZT derivatives against HDAC8 is
summarized in Fig. 3. Taken together, a tetrahydropyrimidine heterocycles (n = 1) is more favorable
than a dihydroimidazole ring and the substitution of the heterycyclic sulfur atom against oxygen has a
detrimental effect. Furthermore, hydrophobic groups are beneficial at positions R1 and R2, whereas
methyl groups at R4 or R5 appear to have negative effects on activity. The R4 position in THPs
tolerates also no exchange of hydrogen against halogens. At position R3 a cyano group is most
beneficial suggesting that a negative partial charge would be required for optimal interaction with
HDAC8.

Fig. 3: Summary of the SAR for the benzothiazine pharmacophore.
2
.4. Selectivity correlates with activity
BTZTs show remarkable potency and selectivity for HDAC8 (Table 2). An increase in potency is
easily achieved by introducing e.g. a general metal chelating group (e.g. hydroxamates) or a larger
hydrophobic group. In such cases the gain in activity is often at the cost of selectivity, because of the
unspecific binding character of such groups. Therefore, it is important to make sure that not only
potency, but also selectivity, is increased during drug optimization. The SAR of BTZT derivatives
shows that the potency of both, the DHI and the THP, series is highly correlated with selectivity for
HDAC8 (Fig. 4). This finding points to a special molecular recognition scheme between BTZTs and
HDAC8, which is different to other HDAC isoenzymes. Since the catalytic center of zinc-dependent
HDACs is highly conserved, the BTZTs are supposed to exploit specific cavities of HDAC8. HDAC8
is known to be exceptionally malleable around the active site binding tunnel. In addition, some crystal
structures show also considerable movement of internal residues that open an additional transient
pocket adjacent to the active site (PDB-ID 1T64) or enable access to the acetate release channel (PDB-
ID 3SFF).

Fig. 4: Selectivity of BTZTs from the DHI- and THP-series. The ratio of logarithmized IC -values
5
0
(
pIC ) against HDAC4 and HDAC8 is plotted versus the pIC of HDAC8 to demonstrate the
50 50
correlation between activity and selectivity. The spearman correlation coefficient r and the p-value are
given for each series.

2
.5. BTZTs bind to the active site pocket
Although BTZTs inhibit the enzyme activity of HDAC8, the binding site would not be necessarily
identical to the active site of the enzyme, since allosteric inhibition would also be possible. To provide
evidence for the site of molecular recognition we exploited a DBD-probe that was developed in our
lab for class IIa HDACs, but also works for HDAC8 . Noncovalent binding of the fluorescent DBD-
probe to HDAC8 leads to increased fluorescence polarization (Fig. 5). After addition of representative
BTZTs 5a-c in 100-fold molar excess, the DBD-probe is displaced from HDAC8 as indicated by a
decrease of the fluorescence polarization signal. Thus, the data provide clear evidence that BTZTs
target the active site of HDAC8.
Fig. 5: Displacement assay demonstrating competitive binding of 5a-c to the active site of HDAC8.
5
00 nM HDAC8 are mixed with 500 nM DBD-probe that leads to an increase in fluorescence
0
polarization (Exc. 420 nm / Em. 520 nm) in assay buffer at 25 C. Upon addition of 50 µM 5a, 5b or
c, the probe is displaced from HDAC8. The free DBD-probe in the absence of HDAC8 is used as low
5
control for complete displacement. Addition of DMSO at the same concentration than that of the
added compound solutions serves as high control for none displacement. Each kinetic curve represents
the average of 6 replicate measurements in order to improve the signal to noise ratio.

2
.6. Docking studies
HDAC8 is known to contain extraordinary flexible regions around the conservative binding pocket
enabling to accommodate a variety of structurally diverse inhibitors. [29, 30] This is documented by
more than 70 crystal structures of HDAC8 (47 from homo sapiens). The inhibitors are supposed to
stabilize the extremely flexible binding site, thereby inducing conformational changes in loops L1 and
L2 flanking the active site binding pocket. HDAC8 essentially undergoes a transition between a more
compact „open“ (e.g. PDB-ID 1T69), a „sub-open“ (e.g. PDB-ID 1T64) with an additional transient
pocket, and a „wide-open“ conformation (e.g. PDB-ID 1VKG). In addition, Whitehead et al. showed a
considerable malleability of HDAC8 at the transition between the conserved binding tunnel and the
acetate release channel in the interior of the protein (e.g. PDB-ID 3SFF) [31]. To elucidate the binding
pose of BTZT’s to HDAC8 we docked compounds 5l, 5c, 5b, 5q and 5d to above mentioned major
conformational representatives of HDAC8 using MOE software. The docking procedure took the
flexibility of residues in proximity to the binding site into account. Subsequently, the resulting protein-
ligand complex was subjected to an energy-minimizing step that was applied to a larger area within a
1
0 Å radius around the bound ligand.
Fig. 6: Molecular docking of 5 BTZT derivatives (5l: red, 5c: green, 5b: orange, 5q: pink, 5d: cyan)
into four different X-ray structures of HDAC8: A) PDB-ID 1T64, B) PDB-ID 1T69, C) PDB-ID
1
VKG, D) PDB-ID 3SFF. E) Docking pose of 5d in PDB-ID 3SFF. Dotted lines indicate protein-
ligand interactions: zinc-sulfur coordination (black), hydrogen-bond (green), π-π-stacking and π-
amide-interaction (magenta) and hydrophobic interactions (orange).
The corresponding energy minimized protein-ligand structures were aligned and superposed for each
conformational representative of HDAC8 in order to analyze structural deviations (Fig. 6A-D). The
various docking poses in Fig. 6A show diverse orientations and are located in the transient secondary
binding pocket far away from the catalytic zinc ion, which is only present in the “sub-open”-
conformation of PDB-ID 1T64. In contrast, all docking poses in the other three crystal structures of

HDAC8 (Fig. 6 B-D) coincide and are located at the transition between the conserved binding pocket
and the acetate release channel. The biggest structural changes of internal amino acid residues in the
crystal structures of 1T69 and 3SFF, which are induced upon binding of 5d, are found in five narrowly
defined regions: L31-R37, W137-A144, S150-D157, D178-D183 and G302-Y306 (Fig. S2). A closer
look at the docking poses of 5d, 5b and 5q in PDB-ID 3SFF reveals several types of interactions:
2
+
Zn -coordination by the thione sulfur, parallel π−π-stacking with W141, a conventional hydrogen
bond with Y306 and amide-π interaction with the peptide backbone of G303 and G304 (Fig. 6E). In
addition, there are several hydrophobic interactions, e.g. with the side chains of K33, I34 and F152.
Docking of the imine-derivative of 5l to PDB-ID 1T69 lead to a very similar binding pose via imine-
nitrogene coordinating to the catalytic zinc ion [27]. The amino acids R37, G303 and W141 that show
significant movements induced by binding of 5d are considered to be involved in gating the acetate
release channel [32]. Moreover, the π-stacking interaction between the aromatic ring of 5d and W141
is very similar to that of known amino-acid derived class I HDAC inhibitors, which are supposed to
achieve isoform selectivity via access to the internal acetate release channel [31]. The hydrophobic
interactions between the methyl-substituent in R1-position with K33, I34 and F152 are in agreement
with the observed increase in activity of this analog, when compared with that of the unsubstituted 5l.
Moreover, the striking beneficial effect of a cyano group in R3-position can also be explained by the
best energy minimized docking pose of 5e in the binding pocket of HDAC8 indicating an additional
hydrogen bond beetwen the nitrogen of the cyano group and the positively charged guanidinium group
of R37 (Fig. S3).
2
.7. Intracellular target engagement
Acetylated SMC3 (ac-SMC3) is a bona fide substrate of HDAC8 and generally accepted as a suitable
marker for HDAC8 enzyme activity [10]. To analyze the efficacy of BTZTs on intracellular HDAC8
activity, we used the SK-N-BE(2)-C neuroblastoma tumor cell line known to depend on HDAC8 for
cell growth and survival [8, 20, 33]. Increasing concentrations of 4b lead to a dose-dependent rise of
the ac-SMC3 level clearly indicating the inhibition of cellular HDAC8 (Fig. 7A/B). The ac-SMC3
level in the presence of 5 µM 4b is about 3.2-fold increased compared with untreated cell control (Fig.
7
B). This compares well with the effect of the known HDAC8 inhibitor PCI-35051 at a concentration
of 6 µM, which increases the ac-SMC3 level by a factor of about 2.9 (Fig. 7A). Thus, BTZTs pass the
cell membrane and bind to their target HDAC8 within cells. The intracellular level of acetylated
histone 4 (ac-H4), which is a substrate of HDACs1, 2 and 3, and acetylated tubulin, a specific
substrate of HDAC6, does not change upon the treatment with up to 5 µM of 4b (Fig. 7C/D).
Therefore, 4b is also isoenzyme selective for HDAC8 within cells in agreement with the biochemical
profiling (Table 2). Furthermore, 5a and 4b reduce clonogenic growth of SK-N-BE(2)-C cells with
comparable or greater efficacy than the reference compound PCI-34051 (Fig. 8). The GI -values of
5
0
the active substances, where 50% of maximal clonogenic growth is inhibited, have been determined by

fitting the data in Fig. 8 yielding values of about 21, 6 and 15 µM for 5a, 4b and PCI-35051,
respectively.
Fig. 7: Inhibition of HDAC8 4b leads to concentration dependent increased acetylation of SMC3 in
SK-N-BE(2)-C neuroblastoma tumor cells. PCI-34051 is a known HDAC8 selective inhibitor and
serves as control. Tubulin and GAPDH are used as loading control, respectively. A) Examplary
quantitave Westernblot of cell extracts in the absence or in the presence at of different concentrations
of 4b. B) The diagram refers to A) and shows mean band intensities and standard deviations from
three independent Westernblots. The y-axis shows relative intensities of ac-SMC3 with respect to an
untreated contol. C) 4b does not interfere with the level of ac-H4 or D) ac-tubulin indicating
isoenzyme selectivity of 4b within tumor cells.
Comparing GI -values with IC -values, there is only a moderate loss of potency when going from the
5
0
50
biochemical to the cell-based assay. For 4b this loss is only a factor of 23. In contrast, the imine-
analogs of the BTZTs appear to be at least 1000-fold less effective on cells compared with
biochemical data [25]. Therefore, our results confirm the initial hypothesis that the dramatic loss in
activity of the chemically instable imine analogs within cells could be prevented by using the GSH
resistant BTZT compounds of this study.

Fig. 8: Colony assay using SK-N-BE(2)-C neuroblastoma tumor cells and treating them with different
concentrations of indicated compounds. Compounds 5a and 4b show comparable and even greater
anti-proliferative activity than the known potent HDAC8 inhibitor PCI-34051. Data are means and
standard deviations of three independent experiments.
2
.8. Drug metabolism and pharmacokinetic properties
The BTZT compounds show also promising in silico drug metabolism and pharmacokinetic (DMPK)
properties (Tab. S2). All compounds fullfill the Lipinski rule of 5, which speaks for good absorption
and permeation. The compounds show a Topological Polar Surface Area (TPSA) [34] between 73 and
2
2
9
7 Å (Tab. S2). These values are significantly below the threshold of 140 Å indicating that the
compounds are supposed to penetrate cell membranes in agreement with the target engagement data
within cells [34]. To prioritize compounds towards a specific target usually normalized activity
measures such as Ligand Efficiency (LE) [35] or Lipophilic Ligand Efficiency (LLE) [36] are
exploited to correct for bias towards lipophilicity and high molecular weight compounds. The great
potential of the BTZT compounds is confirmed by beneficial LE-values ranging from 0.37-0.62 and
LLE-values from 3.1-6.4 (Table S2). Lead compounds usually show LLE-values above 3 and clinical
candidates above 5.
3
. CONCLUSIONS
In our effort to develop potent and selective HDAC8 inhibitors, we identified a novel class of active
substances, BTZTs, that lack classic zinc binding groups, but bind to the active site of the enzyme.
The most potent representative of BTZT derivatives shows double digit nanomolar potency and
considerable isoenzyme selectivity for HDAC8. In contrast to the starting structure, PD-404,182, all
BTZT analogs are extremely robust against millimolar concentration of GSH that occurs in cells and is
responsible for the maintenance of the cellular redox state. BTZTs penetrate cell membranes and
inhibit the enzyme activity of intracellular HDAC8. Furthermore, compound 4b reduces colony
growth of neuroblastoma tumor cells in the range of the known HDAC8 inhibitor PCI-34051 with

hydroxamate zinc binding group. The BTZT pharmacophor bears considerable potential for further
optimization, since both, potency and selectivity, are highly correlated. In addition, BTZT compounds
avoid intrinsic drawbacks of the hydroxamate zinc binding group of conventional HDAC inhibitors
like unselectivity and potential mutagenicity. As potent and selective HDAC8 inhibitors, BTZT
analogs are predestined as tools for the investigation of the biological function of HDAC8. Beyond,
they are promising candidates for the development of anti-cancer drugs and particularly
neuroblastoma.
4
4
3
. EXPERIMENTAL SECTION
.1. Chemical Analytics4-(4,5-Dihydro-1H-imidazol-2-yl)-3-fluorobenzonitrile (2a)
-Fluoro-4-formylbenzonitrile (1a) (1.00 g, 6.71 mmol) was subjected to general procedure, using
ethane-1,2-diamine (493 µL, 443 mg, 7.38 mmol) K CO (2.81 g, 20.3 mmol) and I (2.13 g, 8.38
2
3
2
mmol). 4-(4,5-Dihydro-1H-imidazol-2-yl)-3-fluorobenzonitrile (2a) was received as yellowish solid
(
1.18 g, 6.26 mmol, 93%).
1
H-NMR (CDCl , 500 MHz): δ = 8.16 (t, J = 7.8 Hz, 1H), 7.45 (dd, J = 8.1, 1.5 Hz, 1H), 7.38 (dd, J =
3
1
0.9, 1.5 Hz, 1H), 3.75 (s, 4H).
1
3
C-NMR (CDCl , 126 MHz): δ = 160.90 (s), 159.34 (d, J = 2.2 Hz), 158.88 (s), 132.17 (d, J = 3.4
3
Hz), 128.12 (d, J = 3.2 Hz), 122.90 (d, J = 11.1 Hz), 120.26 (s), 120.04 (s), 116.98 (d, J = 2.0 Hz),
15.33 (d, J = 10.3 Hz).
HPLC/MS (t = 10.551 min, 70 eV, EI) m/z (%) = 190.100 (100) [M+H]
1
+
r
2
2
1
-(2,5-Dibromophenyl)-4,5-dihydro-1H-imidazole (2b)
,5-Dibromobenzaldehyde (1b) (2.38, 9.00 mmol) was subjected to general procedure, using ethane-
,2-diamine (661 µL, 595 mg, 9.90 mmol), K CO (3.73 g, 27.0 mmol) and I (2.86 g, 11.25 mmol). 2-
2
3
2
(
2,5-Dibromophenyl)-4,5-dihydro-1H-imidazole (2b) was received as yellowish solid (2.61 g,
8
.59 mmol, 95%).
1
H-NMR (CDCl , 500 MHz): δ = 7.77 (d, J = 2.4 Hz, 1H), 7.41 (d, J = 8.5 Hz, 1H), 7.35 (dd, J = 8.5,
3
2
.4 Hz, 1H), 3.76 (s, 5H).
1
3
C-NMR (CDCl , 126 MHz): δ = 163.25 (s), 134.74 (s), 134.67 (s), 134.14 (s), 134.10 (s), 121.48 (s),
3
1
19.56 (s), 50.60 (s).
+
GC/MS (t = 19.717 min, 70 eV, EI) m/z (%) = 305.9 (20), 303.9 [M ] (42), 301.9 (25), 276.9 (49),
r
2
74.8 (100), 272.9 (51), 195.9 (18), 193.9 (18), 143.0 (8), 115.0 (18), 88.0 (10) 75.0 (13)
2
-(2-Bromo-6-fluorophenyl)-4,5-dihydro-1H-imidazole (2c)

2
-Bromo-6-fluorobenzaldehyde (1c) (500 mg, 2.46 mmol) was subjected to general procedure, using
ethane-1,2-diamine (181 µL, 163 mg, 2.71 mmol) K CO (1.03 g, 7.46 mmol) and I (781 mg,
2
3
2
3
.08 mmol). 2-(2-Bromo-6-fluorophenyl)-4,5-dihydro-1H-imidazole (2c) was received as yellowish
solid (529 mg, 2.18 mmol, 88%).
1
H-NMR (CDCl , 500 MHz): δ = 7.36 (d, J = 8.1 Hz, 1H), 7.21 (td, J = 8.2, 5.8 Hz, 1H), 7.07 – 7.02
3
(
m, 1H), 3.77 (s, 4H).
1
3
C-NMR (CDCl , 126 MHz): δ = 159.94 (d, J = 407.1 Hz), 159.55 (s), 131.63 (d, J = 9.0 Hz), 128.64
3
(
(
d, J = 3.2 Hz), 122.87 (d, J = 3.4 Hz), 119.20 (dd, J = 864.2, 20.6 Hz), 114.98 (d, J = 22.0 Hz), 62.27
s), 50.56 (s).
+
HPLC/MS (t = 1.269 min, 70 eV, EI) m/z (%) = 245.000/243.100 (100/85) [M+H]
r
2
-(2-Chloro-6-fluorophenyl)-4,5-dihydro-1H-imidazole (2d)
-Chloro-6-fluorobenzaldehyde (1d) (1.00 g, 6.31 mmol) was subjected to general procedure, using
2
ethane-1,2-diamine (463 µL, 417 mg, 6.94 mmol) K CO (2.64 g, 19.1 mmol) and I2 (2.00 g,
2
3
7
.88 mmol). 2-(2-Chloro-6-fluorophenyl)-4,5-dihydro-1H-imidazole (2d) was received as yellowish
solid (1.16 g, 5.82 mmol, 92%).
1
H-NMR (CDCl , 500 MHz): δ = 7.31 – 7.25 (m, 1H), 7.19 (d, J = 8.1 Hz, 1H), 7.03 – 6.98 (m, 1H),
3
3
.76 (s, 1H).
1
3
C-NMR (CDCl , 126 MHz): δ = 160.69 (d, J = 252.5 Hz), 157.51 (d, J = 229.7 Hz), 134.13 (s),
3
1
31.28 (d, J = 9.4 Hz), 125.55 (d, J = 3.1 Hz), 117.73 (dd, J = 666.6, 20.7 Hz), 114.43 (d, J = 22.0
Hz), 62.42 (s), 50.33 (s).
GC/MS (t = 10.062 min, 70 eV, EI) m/z (%) = 198.0 [M ] (30), 171.0 (33), 169.0 (100), 156.0 (6),
+
r
1
34.0 (13), 107.0 (25)
2
-(4-Bromo-2-fluorophenyl)-4,5-dihydro-1H-imidazole (2e)
4
-Bromo-2-fluorobenzaldehyde (1e) (2.03 g, 10.00 mmol) was subjected to general procedure, using
ethane-1,2-diamine (735 µL, 661 mg, 11.0 mmol), K CO (4.15 g, 30.0 mmol) and I2 (3.17 g,
2
3
1
2.5 mmol). 2-(4-Bromo-2-fluorophenyl)-4,5-dihydro-1H-imidazole (2e) was received as yellowish
solid (2.36 g, 9.70 mmol, 97%).
1
H-NMR (CDCl , 500 MHz): δ = 7.91 (t, J = 8.3 Hz, 1H), 7.31 (dd, J = 8.3, 1.8 Hz, 1H), 7.27 (dd, J =
3
1
1.2, 1.8 Hz, 1H), 3.73 (s, 4H).
1
3
C-NMR (CDCl , 126 MHz): δ = 161.34 (s), 160.23 (s), 159.32 (s), 132.04 (d, J = 3.5 Hz), 128.05 (d,
3
J = 3.0 Hz), 125.12 (d, J = 10.2 Hz), 119.80 (d, J = 26.7 Hz), 117.27 (d, J = 10.9 Hz), 49.84 (s).
+
GC/MS (t = 16.687 min, 70 eV, EI) m/z (%) = 244.0 (37), 242 [M ] (40), 215.0 (97), 213.0 (100),
r
1
62.1 (11), 134.0 (23), 107.0 (21), 94.0 (13)
2
-(2-Fluoro-5-iodophenyl)-4,5-dihydro-1H-imidazole (2f)

2
-Fluoro-5-iodobenzaldehyde (1f) (500 mg, 2.00 mmol) was subjected to general procedure, using
ethane-1,2-diamine (147 µL, 132 mg, 2.20 mmol) K CO (838 mg, 6.06 mmol) and I (635 mg,
2
3
2
2
.50 mmol). 2-(2-Fluoro-5-iodophenyl)-4,5-dihydro-1H-imidazole (2f) was received as yellowish
solid (450 mg, 1.55 mmol, 78%).
1
H-NMR (CDCl , 500 MHz): δ = 8.37 (dd, J = 7.0, 2.4 Hz, 1H), 7.66 (ddd, J = 8.7, 4.8, 2.4 Hz, 1H),
3
6
.84 (dd, J = 11.4, 8.7 Hz, 1H), 3.74 (s, 5H).
1
3
C-NMR (CDCl , 126 MHz): δ = 161.62 (s), 159.66 (s), 140.94 (d, J = 8.8 Hz), 139.56 (d, J = 2.5
3
Hz), 120.21 (d, J = 11.7 Hz), 118.44 (s), 118.24 (s), 87.55 (d, J = 2.9 Hz), 49.90 (s).
+
HPLC/MS (t = 2.053 min, 70 eV, EI) m/z (%) = 291.000 (100) [M+H]
r
2
-(2-Bromo-5-fluorophenyl)-4,5-dihydro-1H-imidazole (2g)
2
-Bromo-5-fluorobenzaldehyde (1g) (2.03 g, 10.00 mmol) was subjected to general procedure, using
ethane-1,2-diamine (735 µL, 661 mg, 11.0 mmol), K CO (4.15 g, 30.0 mmol) and I2 (3.17 g,
2
3
1
2.5 mmol). 2-(2-Bromo-5-fluorophenyl)-4,5-dihydro-1H-imidazole (2g) was received as yellowish
solid (2.13 g, 8.76 mmol, 88%).
1
H-NMR (CDCl , 500 MHz): δ = 7.51 (dd, J = 8.8, 5.1 Hz, 1H), 7.36 (dd, J = 8.9, 3.1 Hz, 1H), 6.96
3
(
ddd, J = 8.8, 7.7, 3.1 Hz, 1H), 3.75 (s, 1H).
1
3
C-NMR (CDCl , 126 MHz): δ = 163.46 (s), 161.67 (d, J = 248.5 Hz), 134.79 (d, J = 7.8 Hz), 134.79
3
–
134.58 (m), 118.55 (d, J = 2.4 Hz), 118.36 (s), 115.10 (d, J = 3.2 Hz), 50.60 (s).
+
GC/MS (t = 16.301 min, 70 eV, EI) m/z (%) = 244.0 (33), 242 [M ] (49), 214.9 (98), 212.9 (100),
r
1
62.0 (13), 134.0 (52), 107.0 (45), 94.0 (15)
2
-(2,6-Difluoro-4-bromoophenyl)-4,5-dihydro-1H-imidazole (2h)
2
,6-Difluoro-4-bromobenzaldehyde (1h) (500 mg, 2.26 mmol) was subjected to general procedure,
using ethane-1,2-diamine (166 µL, 150 mg, 2.49 mmol), K CO (947 mg, 6.86 mmol) and I (718 mg,
2
3
2
2
.83 mmol). 2-(2,6-Difluoro-4-bromoophenyl)-4,5-dihydro-1H-imidazole (2h) was received as
yellowish solid (273 mg, 1.05 mmol, 46%).
1
H-NMR (CDCl , 300 MHz): δ = 7.18 (s, 1H), 7.15 (s, 1H), 3.80 (s, 4H).
3
2
-(2-Fluoro-5-methylphenyl)-4,5-dihydro-1H-imidazole (2i)
2
-Fluoro-5-methylbenzaldehyde (1i) (1.38 g, 10.00 mmol) was subjected to general procedure, using
ethane-1,2-diamine (734 µL, 660 mg, 11.0 mmol), K CO (4.18 g, 30.3 mmol) and I2 (3.17 g,
2
3
1
2.5 mmol). 2-(2-Fluoro-5-methylphenyl)-4,5-dihydro-1H-imidazole (2i) was received as yellowish
solid (1.71 g, 9.57 mmol, 96%).
1
H-NMR (CDCl , 500 MHz): δ = 7.82 (dd, J = 7.2, 2.1 Hz, 1H), 7.17 – 7.11 (m, 1H), 6.93 (dd, J =
3
1
1.7, 8.4 Hz, 1H), 5.69 (s, 1H), 3.72 (s, 5H), 2.27 (s, 4H).

1
3
C-NMR (CDCl , 126 MHz): δ = 161.30 (s), 159.03 (d, J = 247.1 Hz), 134.12 (s), 133.72 – 130.23
3
(
m), 117.14 (d, J = 10.8 Hz), 115.80 (d, J = 23.5 Hz), 49.62 (s), 20.45 (s).
2
-(2-Fluoro-4-chloroophenyl)-4,5-dihydro-1H-imidazole (2j)
2
-Fluoro-4-chlorobenzaldehyde (1j) (500 mg, 3.15 mmol) was subjected to general procedure, using
ethane-1,2-diamine (232 µL, 208 mg, 3.47 mmol), K CO (1.32 g, 9.55 mmol) and I (1.00 g,
2
3
2
3
.94 mmol). 2-(2-Fluoro-4-chloroophenyl)-4,5-dihydro-1H-imidazole (2j) was received as yellowish
solid (590 mg, 2.97 mmol, 92%).
1
H-NMR (CDCl , 300 MHz): δ = 7.37 (dd, J = 14.4, 6.7 Hz, 1H), 7.01 (t, J = 7.9 Hz, 1H), 3.36 (t, J =
3
5
.7 Hz, 2H), 1.83 (dt, J = 11.3, 5.7 Hz, 1H).
1
3
C-NMR (CDCl , 126 MHz): δ = 155.07 (d, J = 252.6 Hz), 152.64 (s), 132.31 (s), 128.61 (d, J = 1.5
3
Hz), 124.74 (d, J = 4.5 Hz), 123.59 (d, J = 12.8 Hz), 121.51 (d, J = 18.3 Hz), 52.96 (d, J = 981.5 Hz),
4
0.96 (s), 19.54 (s).
2
-(2-Fluoro-6-methylphenyl)-4,5-dihydro-1H-imidazole (2k)
2
-Fluoro-6-methylbenzaldehyde (1k) (1.08 g, 7.85 mmol) was subjected to general procedure, using
ethane-1,2-diamine (531 µL, 479 mg, 7.96 mmol), K CO (3.00 g, 21.7 mmol) and I2 (2.30 g,
2
3
9
.05 mmol). 2-(2-Fluoro-6-methylphenyl)-4,5-dihydro-1H-imidazole (2k) was received as yellowish
solid (1.24 g, 6.95 mmol, 89%).
GC/MS (t = 14.195 min, 70 eV, EI) m/z (%) = 178.1 (87), 149.1 (65), 148.1 (100), 122.0 (32), 101.0
r
(
14), 28.0 (40)
2
-(2-Fluoro-6-iodophenyl)-4,5-dihydro-1H-imidazole (2l)
2
-Fluoro-6-iodobenzaldehyde (1l) (500 mg, 2.00 mmol) was subjected to general procedure, using
ethane-1,2-diamine (147 µL, 132 mg, 2.20 mmol), K CO (838 mg, 6.06 mmol) and I (635 mg,
2
3
2
2
.50 mmol). 2-(2-Fluoro-6-iodophenyl)-4,5-dihydro-1H-imidazole (2l) was received as yellowish
solid (575 mg, 1.98 mmol, 99%).
1
H-NMR (CDCl , 500 MHz): δ = 7.61 (ddd, J = 6.7, 4.0, 1.0 Hz, 1H), 7.08 – 7.07 (m, 1H), 7.07 – 7.04
3
(
m, 1H), 3.78 (s, 4H).
1
3
C-NMR (CDCl , 126 MHz): δ = 161.18 (d, J = 121.0 Hz), 159.87 (d, J = 300.5 Hz), 136.05 – 134.48
3
(
m), 132.16 (d, J = 8.6 Hz), 121.44 (dd, J = 1208.4, 20.2 Hz), 115.77 (d, J = 21.9 Hz), 96.73 (d, J =
1
.0 Hz), 62.01 (s), 50.59 (s).
+
HPLC/MS (t = 1.415 min, 70 eV, EI) m/z (%) = 291.000 (100) [M+H]
r
2
-(3-Chloro-2-fluorophenyl)-4,5-dihydro-1H-imidazole (2m)
-Fluoro-3-chlorobenzaldehyde (1m) (500 mg, 3.15 mmol) was subjected to general procedure, using
2
ethane-1,2-diamine (232 µL, 208 mg, 3.47 mmol), K CO (1.32 g, 9.55 mmol) and I (1.00 g,
2
3
2

3
.94 mmol). 2-(3-Chloro-2-fluorophenyl)-4,5-dihydro-1H-imidazole (2m) was received as yellowish
solid (510 mg, 2.55 mmol, 81%).
1
H-NMR (CDCl , 300 MHz): δ = 7.87 (t, J = 7.2 Hz, 1H), 7.41 (t, J = 7.5 Hz, 1H), 7.07 (t, J = 7.9 Hz,
3
1
H), 5.44 (s, 1H), 3.72 (s, 4H).
1
3
C-NMR (CDCl , 75 MHz): δ = 160.24 (s), 156.10 (d, J = 252.2 Hz), 132.56 (s), 129.23 (d, J = 2.3
3
Hz), 124.75 (d, J = 4.3 Hz), 121.83 (d, J = 19.1 Hz), 119.75 (d, J = 11.0 Hz), 49.85 (s).
+
HPLC/MS (t = 1.689 min, 70 eV, EI) m/z (%) = 199.000 (100) [M+H] , 201.100 (32)
r
2
-(2,3-Difluorophenyl)-4,5-dihydro-1H-imidazole (2n)
2
,3-Difluorobenzaldehyde (1n) (1.43 g, 10.00 mmol) was subjected to general procedure, using
ethane-1,2-diamine (738 µL, 664 mg, 11.0 mmol), K CO (4.21 g, 30.4 mmol) and I2 (3.19 g,
2
3
1
2.5 mmol). 2-(2,3-Difluorophenyl)-4,5-dihydro-1H-imidazole (2n) was received as yellowish solid
(
1.55 g, 8.51 mmol, 85%).
1
H-NMR (CDCl , 500 MHz): δ = 7.75 (ddt, J = 8.0, 6.3, 1.7 Hz, 1H), 7.20 (dtd, J = 9.8, 8.0, 1.7 Hz,
3
1
H), 7.08 (tdd, J = 8.2, 4.8, 1.6 Hz, 1H), 5.15 (s, 1H), 3.74 (d, J = 7.7 Hz, 4H).
1
3
C-NMR (CDCl , 126 MHz): δ = 160.13 (s), 151.79 (d, J = 13.6 Hz), 150.21 (d, J = 14.3 Hz), 149.82
3
(
d, J = 13.6 Hz), 148.20 (d, J = 14.2 Hz), 125.46 (s), 124.33 (dd, J = 6.3, 4.7 Hz), 120.31 (d, J = 7.7
Hz), 119.07 (d, J = 17.1 Hz), 49.91 (s).
GC/MS (t = 13.551 min, 70 eV, EI) m/z (%) = 182.1 [M ] (38), 153.0 (100), 126.0 (9), 113.0 (9)
+
r
3
-Bromo-4-(4,5-dihydro-1H-imidazol-2-yl)-N,N-dimethylaniline (2o)
2
-Bromo-4-(dimethylamino)benzaldehyde (1o) (458 mg, 2.01 mmol) was subjected to general
procedure, using ethane-1,2-diamine (147 µL, 133 mg, 2.21 mmol), K CO (841 mg, 6.08 mmol) and
2
3
I2 (637 mg, 2.51 mmol). 3-Bromo-4-(4,5-dihydro-1H-imidazol-2-yl)-N,N-dimethylaniline (2o) was
received as yellowish solid (520 mg, 1.94 mmol, 97%).
1
1
1
H-NMR (CDCl , 500 MHz): δ = H-NMR (CDCl , 500 MHz): δ = H-NMR (CDCl , 500 MHz):
3
3
3
1
δ = H-NMR (CDCl , 500 MHz): δ = 7.60 (d, J = 8.8 Hz, 1H), 6.78 (d, J = 2.6 Hz, 1H), 6.58 (dd, J =
3
8
.8, 2.6 Hz, 1H), 3.71 (s, 4H), 2.94 (d, J = 2.9 Hz, 7H).
1
3
C-NMR (CDCl , 126 MHz): δ = 164.68 (s), 151.90 (s), 132.14 (s), 121.72 (s), 118.69 (s), 115.79 (s),
3
1
10.76 (s), 50.05 (s), 40.05 (s).
+
GC/MS (t = 23.206 min, 70 eV, EI) m/z (%) = 269.1 (80), 267.1 [M ] (85), 240.1 (97), 238.1 (100),
r
2
25.1 (18), 187.2 (9), 159.1 (15), 143.1 (13), 119.2 (13)
2
-(2-Fluoro-5-methoxyphenyl)-4,5-dihydro-1H-imidazole (2p)
2
-Fluoro-5-methoxybenzaldehyde (1p) (500 mg, 3.24 mmol) was subjected to general procedure,
using ethane-1,2-diamine (238 µL, 214 mg, 3.57 mmol), K CO (1.36 g, 9.83 mmol) and I (1.03 g,
2
3
2

4
.05 mmol). 2-(2-Fluoro-5-methoxyphenyl)-4,5-dihydro-1H-imidazole (2p) was received as yellowish
solid (876 mg, 3.38 mmol, 84%).
1
H-NMR (DMSO-d , 300 MHz): δ = 7.85 – 7.76 (m, 1H), 7.35 – 7.18 (m, 2H), 4.22 – 3.86 (m, 8H).
6
+
HPLC/MS (t = 1.597 min, 70 eV, EI) m/z (%) = 195.100 (100) [M+H]
r
2
-(2-Fluoro-4-methoxyphenyl)- 4,5-dihydro-1H-imidazole (2q)
2
-Fluoro-4-methoxybenzaldehyde (1x) (308 mg, 2.00 mmol) was subjected to general procedure,
using ethane-1,2-diamine (147 µL, 132 mg, 2.20 mmol), K CO (828 mg, 5.99 mmol) and I (639 mg,
2
3
2
2
.52 mmol). 2-(2-Fluoro-4-methoxyphenyl)-4,5-dihydro-1H-imidazole (2q) was received as brown
solid (409 mg, 1.97 mmol, 98%).
1
H NMR (300 MHz, CDCl ) δ 7.95 (t, J = 8.9 Hz, 1H), 6.69 (dd, J = 8.8, 2.5 Hz, 1H), 6.56 (dd, J =
3
1
3.8, 2.5 Hz, 1H), 3.77 (s, 3H), 3.69 (s, 4H).
1
3
C-NMR (CDCl , 75 MHz): δ = 163.25 (s), 162.68 (d, J = 12.1 Hz), 160.89 (s), 159.95 (s), 131.67 (d,
3
J = 4.9 Hz), 110.61 (s), 101.71 (s), 101.35 (s), 55.66 (s), 49.59 (s).
GC/MS (t = 18.404 min, 70 eV, EI) m/z (%) = 194.200 (50), 165.100 (100), 150.1 (37), 95.1 (9)
r
2
-(2-Bromo-5-chlorophenyl)-4,5-dihydro-1H-imidazole (2r)
2
-Bromo-5-chlorobenzaldehyde (1r) (878 mg, 4.00 mmol) was subjected to general procedure, using
ethane-1,2-diamine (294 µL, 265 mg, 4.40 mmol), K CO (1.66 g, 12.0 mmol) and I2 (1.27 g,
2
3
5
.00 mmol). 2-(2-Bromo-5-chlorophenyl)-4,5-dihydro-1H-imidazole (2r) was received as yellowish
solid (876 mg, 3.38 mmol, 84%).
1
H-NMR (CDCl , 500 MHz): δ = 7.61 (d, J = 2.6 Hz, 1H), 7.47 (d, J = 8.6 Hz, 1H), 7.20 (dd, J = 8.6,
3
2
.6 Hz, 1H), 3.74 (s, 4H).
1
3
C-NMR (CDCl , 126 MHz): δ = 163.28 (s), 134.39 (s), 133.64 (s), 131.15 (s), 131.07 (s), 118.69 (s),
3
5
0.55 (s).
+
GC/MS (t = 18.778 min, 70 eV, EI) m/z (%) = 260.000 [M ] (42), 258.0 (33), 231.0 (100), 229.0 (78),
r
1
50.1 (23), 123.1 (16) 75.1 (10)
2
-(2,6-Difluorophenyl)-4,5-dihydro-1H-imidazole (2s)
2
,6-Difluorobenzaldehyde (1s) (1.42 g, 10.00 mmol) was subjected to general procedure, using
ethane-1,2-diamine (734 µL, 660 mg, 11.0 mmol), K CO (4.15 g, 30.0 mmol) and I (3.17 g,
2
3
2
1
2.5 mmol). 2-(2,6-Difluorophenyl)-4,5-dihydro-1H-imidazole (2s) was received as yellowish solid
(
1.54 g, 8.48 mmol, 85%).
1
H-NMR (CDCl , 500 MHz): δ = 7.30 (tt, J = 8.4, 6.3 Hz, 1H), 6.93 – 6.87 (m, 2H), 5.03 (s, 1H), 3.73
3
(
s, 4H).
1
3
C-NMR (CDCl , 126 MHz): δ = 160.96 (dd, J = 253.6, 6.5 Hz), 156.37 (s), 131.41 (t, J = 10.7 Hz),
3
1
12.00 (dd, J = 21.6, 4.1 Hz), 109.57 (t, J = 17.3 Hz), 50.08 (s).

+
GC/MS (t = 13.858 min, 70 eV, EI) m/z (%) = 182.100 [M ] (32), 153.100 (100)
r
2
-(2-Fluoro-4-methylphenyl)-4,5-dihydro-1H-imidazole (2t)
2
-Fluoro-4-methylbenzaldehyde (1t) (1.38 g, 10.00 mmol) was subjected to general procedure, using
ethane-1,2-diamine (734 µL, 660 mg, 11.0 mmol), K CO (4.18 g, 30.3 mmol) and I (3.17 g,
2
3
2
1
2.5 mmol). 2-(2-Fluoro-4-methylphenyl)-4,5-dihydro-1H-imidazole (2t) was received as yellowish
solid (1.68 g, 9.43 mmol, 94%).
1
H-NMR (CDCl , 500 MHz): δ = 7.89 (t, J = 8.0 Hz, 1H), 6.96 (d, J = 8.0 Hz, 1H), 6.87 (d, J = 12.8
3
Hz, 1H), 5.80 (s, 1H), 3.73 (s, 4H), 2.33 (s, 3H).
1
3
C-NMR (CDCl , 126 MHz): δ = 161.21 (s, J = 1.7 Hz), 160.66 (d, J = 249.7 Hz), 143.65 (d, J = 9.0
3
Hz), 130.63 (d, J = 3.0 Hz), 125.44 (d, J = 2.1 Hz), 116.54 (d, J = 23.1 Hz), 114.66 (d, J = 10.8 Hz),
4
9.46 (s), 21.32 (s).
2
2
1
-(2-Bromophenyl)-4,5-dihydro-1H-imidazole (2u)
-Bromobenzaldehyde (1u) (924 mg, 5.00 mmol) was subjected to general procedure, using ethane-
,2-diamine (367 µL, 330 mg, 5.50 mmol), K CO (2.09 g, 15.1 mmol) and I (1.58 g, 6.24 mmol). 2-
2
3
2
(
2-Bromophenyl)-4,5-dihydro-1H-imidazole (2u) was received as yellowish solid (926 mg,
4
.11 mmol, 82%).
1
H-NMR (CDCl , 300 MHz): δ = 7.55 (td, J = 7.9, 1.6 Hz, 2H), 7.25 (tdd, J = 15.2, 7.5, 1.6 Hz, 2H),
3
3
.72 (s, 4H).
1
3
C-NMR (CDCl , 75 MHz): δ = 164.50 (s), 133.21 (s), 132.73 (s), 131.09 (d, J = 3.3 Hz), 127.38 (s),
3
1
20.80 (s), 50.30 (s).
+
GC/MS (t = 16.819 min, 70 eV, EI) m/z (%) = 224.100 [M ] (48),
r
2
-(2-Fluoro-3-methylphenyl)-4,5-dihydro-1H-imidazole (2v)
2
-Fluoro-3-methylbenzaldehyde (1v) (500 mg, 3.62 mmol) was subjected to general procedure, using
ethane-1,2-diamine (266 µL, 239 mg, 3.98 mmol), K CO (1.52 g, 10.97 mmol) and I (1.15 g,
2
3
2
4
.52 mmol). 2-(2-Fluoro-3-methylphenyl)-4,5-dihydro-1H-imidazole (2v) was received as yellowish
solid (503 mg, 2.82 mmol, 78%).
1
H NMR (300 MHz, DMSO) δ 7.50 (t, J = 7.2 Hz, 1H), 6.98 (t, J = 7.3 Hz, 1H), 6.78 (t, J = 7.6 Hz,
1
H), 4.33 (s, 2H), 3.46 (s, 4H).
1
3
C-NMR (DMSO, 75 MHz): δ = 160.83 (s), 158.55 (d, J = 249.6 Hz), 133.09 (d, J = 5.8 Hz), 134.07
127.87 (m), 127.65 (d, J = 2.7 Hz), 124.94 (d, J = 18.7 Hz), 123.30 (d, J = 3.9 Hz), 49.07 (s), 13.99
–
(
d, J = 5.2 Hz).
2
-(2-Fluoro-5-methylphenyl)-1,4,5,6-tetrahydropyrimidine (3a)

2
-Fluoro-5-methylbenzaldehyde (1i) (500 mg, 3.62 mmol) was subjected to general procedure, using
propane-1,3-diamine (328 µl, 295 mg, 3.98 mmol), K CO (1.52 g, 11.0 mmol) and I (1.15 g,
2
3
2
4
.52 mmol). 2-(2-Fluoro-5-methylphenyl)-1,4,5,6-tetrahydropyrimidine (3a) was received as brown
solid (800 mg, 3.11 mmol, 86%).
1
H-NMR (CDCl , 500 MHz): δ = 7.33 (dd, J = 6.8, 1.8 Hz, 1H), 7.18 – 7.12 (m, 1H), 6.89 (dd, J =
3
1
0.5, 8.5 Hz, 1H), 3.34 (t, J = 5.8 Hz, 4H), 2.23 (s, 3H), 1.84 – 1.76 (m, 2H).
1
3
C-NMR (CDCl , 126 MHz): δ = 157.71 (d, J = 249.4 Hz), 154.52 (s), 134.26 (d, J = 3.1 Hz), 133.53
3
(
(
d, J = 8.3 Hz), 130.39 (s), 119.32 (d, J = 12.3 Hz), 115.95 (d, J = 22.0 Hz), 40.26 (s), 20.47 (s), 19.01
s).
+
HPLC/MS (t = 1.728 min, 70 eV, EI) m/z (%) = 193.100 (100) [M+H]
r
2
-(2-Fluoro-6-iodophenyl)-1,4,5,6-tetrahydropyrimidine (3b)
2
-Fluoro-6-iodobenzaldehyde (500 mg, 2.00 mmol) (1l) was subjected to general procedure, using
propane-1,3-diamine (181 µl, 163 mg, 2.20 mmol), K CO (838 mg, 6.06 mmol) and I (635 mg,
2
3
2
2
.50 mmol). 2-(2-Fluoro-6-iodophenyl)-1,4,5,6-tetrahydropyrimidine (3b) was received as brown
solid (268 mg, 0.88 mmol, 44%).
+
HPLC/MS (t = 1.400 min, 70 eV, CI) m/z (%) = 305.000 (100) [M+H]
r
2
-(2-Chloro-6-fluorophenyl)-1,4,5,6-tetrahydropyrimidine (3c)
-Chloro-6-fluorobenzaldehyde (1d) (1.00 g, 6.31 mmol) was subjected to general procedure, using
2
propane-1,3-diamine (584 µl, 514 mg, 6.94 mmol), K CO (2.64 g, 19.1 mmol) and I (2.00 g,
2
3
2
7
.88 mmol). 2-(2-Chloro-6-fluorophenyl)-1,4,5,6-tetrahydropyrimidine (3c) was received as brown
solid (1.24 g, 5.82 mmol, 92%).
1
H-NMR (CDCl , 500 MHz): δ = 7.19 (td, J = 8.3, 6.0 Hz, 1H), 7.00 (d, J = 8.1 Hz, 1H), 6.83 (td, J =
3
8
.5, 0.7 Hz, 1H), 3.01 (t, J = 5.8 Hz, 4H), 1.61 – 1.54 (m, 2H).
1
3
C-NMR (CDCl , 126 MHz): δ = 159.19 (d, J = 253.0 Hz), 151.43 (s), 132.80 (s), 131.73 (d, J = 9.2
3
Hz), 124.86 (s), 120.84 (d, J = 19.6 Hz), 113.94 (d, J = 21.3 Hz), 18.35 (s).
+
HPLC/MS (t = 1.164 min, 70 eV, EI) m/z (%) = 213.100/215.100 (100) [M+H]
r
2
-(2-Fluoro-6-methylphenyl)-1,4,5,6-tetrahydropyrimidine (3d)
2
-Fluoro-6-methylbenzaldehyde (1k) (1.01 g, 7.30 mmol) was subjected to general procedure, using
propane-1,3-diamine (663 µl, 590 mg, 7.75 mmol), K CO (2.99 g, 21.7 mmol) and I (2.30 g,
2
3
2
9
.06 mmol). 2-(2-Fluoro-4-methylphenyl)-1,4,5,6-tetrahydropyrimidine (3d) was received as
yellowish solid (1.13 mg, 5.87 mmol, 80%).
GC/MS (t = 15.93 min, 70 eV, EI) m/z (%) = 192 [M ] (100), 177 (45), 173 (19), 135 (46), 116 (15),
+
r
8
9 (12).

1
H-NMR (CDCl , 300 MHz): δ = 7.25 (s, J = 3.7 Hz, 1H), 7.17 (td, J = 8.0, 5.9 Hz, 1H), 6.90 (d, J =
3
7
.6 Hz, 1H), 6.82 (t, J = 8.8 Hz, 1H), 3.12 – 3.02 (m, 4H), 2.17 (s, 3H), 1.71 – 1.60 (m, 2H).
1
3
C-NMR (CDCl , 75 MHz): δ = 159.7 (d, J = 247.3 Hz), 152.4 (s), 138.7 (d, J = 2.4 Hz), 130.3 (d, J =
3
8
.8 Hz), 125.8 (d, J = 2.8 Hz), 123.9 (d, J = 16.5 Hz), 112.8 (d, J = 21.6 Hz), 40.8 (s), 19.9 (s), 18.6 (d,
J = 2.1 Hz).
4
-Fluoro-3-(1,4,5,6-tetrahydropyrimidin-2-yl)benzonitrile (3e)
4
-Fluoro-3-formylbenzonitrile (1a) (1.00 g, 6.71 mmol) was subjected to general procedure, using
propane-1,3-diamine (608 µl, 547 mg, 7.38 mmol), K CO (2.81 g, 20.3 mmol) and I (2.13 g,
2
3
2
8
.38 mmol). 4-Fluoro-3-(1,4,5,6-tetrahydropyrimidin-2-yl)benzonitrile (3e) was received as brown
solid (879 mg, 4.33 mmol, 65%).
+
HPLC/MS (t = 11.075 min, 70 eV, EI) m/z (%) = 204.100 (100) [M+H]
r
2
-(2-Bromo-5-chlorophenyl)-1,4,5,6-tetrahydropyrimidine (3f)
-Bromo-5-chlorobenzaldehyde (1r) (878 g, 4.00 mmol) was subjected to general procedure, using
2
propane-1,3-diamine (367 ml, 327 mg, 4.41 mmol), K CO (1.66 g, 12.0 mmol) and I (1.30 g,
2
3
2
5
.12 mmol). 2-(2-Bromo-5-chlorophenyl)-1,4,5,6-tetrahydropyrimidine (3f) was received as yellow
solid (992.7 mg, 3.63 mmol, 91%).
1
H-NMR (CDCl , 500 MHz): δ = 7.42 (d, J = 8.5 Hz, 12H), 7.36 (d, J = 2.6 Hz, 11H), 7.28 – 7.24 (m,
3
2
H), 7.14 (dd, J = 8.5, 2.6 Hz, 12H), 4.97 (s, 13H), 3.40 – 3.33 (m, 50H), 1.84 – 1.75 (m, 25H).
1
3
C-NMR (CDCl , 125 MHz): δ = 154.4, 140.4, 134.0, 133.5, 130.4, 130.2, 118.7, 42.2, 20.5.
3
2
-(2-Bromo-4-(trifluoromethyl)phenyl)-1,4,5,6-tetrahydropyrimidine (3g)
2
-Bromo-4-(trifluoromethyl)benzaldehyde (1v) (504 mg, 1.99 mmol) was subjected to general
procedure, using propane-1,3-diamine (181 µl, 163 mg, 2.19 mmol), K CO (835 mg, 6.04 mmol) and
2
3
I2 (632 mg, 2.49 mmol). 2-(2-Bromo-4-(trifluoromethyl)phenyl)-1,4,5,6-tetrahydropyrimidine (3g)
was received as brown solid (478 mg, 1.56 mmol, 78%).
1
H-NMR (CDCl , 500 MHz): δ = 7.64 (t, J = 5.4 Hz, 2H), 7.43 (dd, J = 8.3, 2.0 Hz, 1H), 3.42 – 3.35
3
(
m, 4H), 1.87 – 1.79 (m, 2H).
1
3
C-NMR (CDCl , 126 MHz): δ = 154.60 (s), 139.47 (s), 133.71 (s), 130.26 (s), 129.99 (s), 127.35 (d,
3
J = 3.5 Hz), 126.94 (d, J = 3.2 Hz), 125.02 (s), 42.15 (s), 30.98 (s), 20.43 (s).
+
GC/MS (t = 17.461 min, 70 eV, EI) m/z (%) = 307.000 [M ] (78),
r
2
-(2-Fluoro-4-methylphenyl)-1,4,5,6-tetrahydropyrimidine (3h)
2
-Fluoro-4-methylbenzaldehyde (1t) (1.38 g, 10.0 mmol) was subjected to general procedure, using
propane-1,3-diamine (916 µl, 815 mg, 11.0 mmol), K CO (4.15 g, 30.0 mmol) and I (3.19 g,
2
3
2

1
2.6 mmol). 2-(2-Fluoro-4-methylphenyl)-1,4,5,6-tetrahydropyrimidine (3h) was received as brown
viscous oil (1.93 mg, 10.0 mmol, quant.).
GC/MS (t = 16.55 min, 70 eV, EI) m/z (%) = 192 [M ] (65), 191 [(M-H) ] (100), 173 (38), 136 (68),
+
+
r
1
16 (20), 89 (18).
1
H-NMR (CDCl , 500 MHz): δ = 7.48 (t, J = 8.0 Hz, 1H), 6.93 – 6.87 (m, 1H), 6.81 (d, J = 12.3 Hz,
3
1
H), 6.36 (s, 1H), 3.38 (dd, J = 13.1, 7.3 Hz, 4H), 2.31 (s, 3H), 1.85 – 1.74 (m, 2H).
1
3
C-NMR (CDCl , 125 MHz): δ = 159.8 (d, J = 248.8 Hz), 153.0 (s), 142.8 (d, J = 8.6 Hz), 130.1 (d, J
3
=
2.6 Hz), 125.2 (d, J = 2.1 Hz), 119.7 (d, J = 11.8 Hz), 116.5 (d, J = 22.7 Hz), 41.4 (s), 21.2 (s), 20.1
(
s).
2
-(2-Fluoro-5-iodophenyl)-1,4,5,6-tetrahydropyrimidine (3i)
2
-Fluoro-5-iodobenzaldehyde (1f) (500 mg, 2.00 mmol) was subjected to general procedure, using
propane-1,3-diamine (181 µl, 163 mg, 2.20 mmol), K CO (838 mg, 6.06 mmol) and I (635 mg,
2
3
2
2
.50 mmol). 2-(2-Fluoro-5-iodophenyl)-1,4,5,6-tetrahydropyrimidine (3i) was received as brown solid
(
433 mg, 1.42 mmol, 71%).
1
H-NMR (CDCl , 500 MHz): δ = 8.02 (dd, J = 7.0, 2.4 Hz, 1H), 7.60 (ddd, J = 8.6, 4.7, 2.4 Hz, 1H),
3
6
.79 (dd, J = 11.1, 8.7 Hz, 1H), 3.46 – 3.40 (m, 4H), 1.87 – 1.79 (m, 2H).
1
3
C-NMR (CDCl , 126 MHz): δ = 160.00 (d, J = 249.5 Hz), 150.83 (s), 140.05 (d, J = 8.5 Hz), 139.18
3
(
d, J = 2.4 Hz), 125.88 (d, J = 12.7 Hz), 118.29 (d, J = 24.5 Hz), 87.52 (s), 20.42 (s).
+
HPLC/MS (t = 3.402 min, 70 eV, EI) m/z (%) = 305.000 (100) [M+H]
r
2
-(4-Bromo-2-fluorophenyl)-1,4,5,6-tetrahydropyrimidine (3j)
4
-Bromo-2-fluorobenzaldehyde (1e) (5.08 g, 25.0 mmol) was subjected to general procedure, using
propane-1,3-diamine (2.29 ml, 2.04 g, 27.5 mmol), K CO (10.4 g, 75.0 mmol) and I (7.98 g,
2
3
2
3
1.4 mmol). 2-(4-Bromo-2-fluorophenyl)-1,4,5,6-tetrahydropyrimidine (3j) was received as brown
solid (800 mg, 3.11 mmol, 12%).
1
H-NMR (CDCl , 500 MHz): δ = 7.61 (t, J = 8.3 Hz, 1H), 7.25 (dd, J = 8.4, 1.9 Hz, 1H), 7.20 (dd, J =
3
1
0.9, 1.8 Hz, 1H), 5.38 (s, 1H), 3.43 (t, J = 5.8 Hz, 4H), 1.81 (p, J = 5.8 Hz, 2H).
1
3
C-NMR (CDCl , 125 MHz): δ = 159.78 (d, J = 252.3 Hz), 151.12 (s), 131.76 (d, J = 3.6 Hz), 127.79
3
(
d, J = 3.0 Hz), 123.76 (d, J = 10.0 Hz), 123.29 (d, J = 11.7 Hz), 119.55 (d, J = 26.9 Hz), 42.13 (s),
2
0.53 (s).
2
-(2,4-Difluorophenyl)-1,4,5,6-tetrahydropyrimidine (3k)
2
,4-Difluorobenzaldehyde (1w) (1.43 g, 10.0 mmol) was subjected to general procedure, using
propane-1,3-diamine (916 µL, 815 mg, 11.0 mmol), K CO (4.15 g, 30.0 mmol) and I (3.18 g,
2
3
2
1
2.5 mmol). 2-(2,4-Difluorophenyl)-1,4,5,6-tetrahydropyrimidine (3k) was received as brown solid
(
1.73 g, 8.80 mmol, 88%).