Synthesis and biological evaluation of a new class of benzothiazines as neuroprotective agents

European Journal of Medicinal Chemistry 126 (2017) 614e630

Contents lists available at ScienceDirect

European Journal of Medicinal Chemistry

journal homepage: h ttp://www.elsevier.com/locate/ejmech

Research paper

Synthesis and biological evaluation of a new class of benzothiazines as

neuroprotective agents

,

Alessandra Mancini ^a, Alessia Chelini ^a, Angela Di Capua ^{a b}, Loretta Castelli ^c,

Simone Brogi ^a, Marco Paolino ^a, Germano Giuliani ^a, Andrea Cappelli ^a, Maria Frosini ^d,

Lorenzo Ricci ^d, Erminia Leonelli ^d, Gianluca Giorgi ^a, Antonio Giordani ^e,

,

*

Jacopo Magistretti ^c, Maurizio Anzini ^a

a

ꢀ

Dipartimento di Biotecnologie, Chimica e Farmacia, Universita di Siena, Via A. Moro, 53100 Siena, Italy

^bEskitis Institute for Drug Discovery, Grifﬁth University, Brisbane, Queensland, Australia

c

ꢀ

Dipartimento di Biologia e Biotecnologie, Fisiologia Generale, Universita di Pavia, Via Forlanini 6, 27100 Pavia, Italy

Dipartimento di Scienze della Vita, Universita di Siena, Via A. Moro, 53100 Siena, Italy

d

ꢀ

^eRottapharm Biotech, Via Valosa di Sopra 7, 20052 Monza, Italy

a r t i c l e i n f o

a b s t r a c t

Article history:

Neurodegenerative diseases are disorders related to the degeneration of central neurons that gradually

lead to various, severe alterations of cognitive and/or motor functions. Currently, for no such diseases

does any pharmacological treatment exist able to arrest its progression. Riluzole (1) is a small molecule

able to interfere with multiple cellular and molecular mechanisms of neurodegeneration, and is the only

approved treatment of amyotrophic lateral sclerosis (ALS), the progression of which proved to signiﬁ-

cantly slow, thus increasing somewhat average survival. Here we report the synthesis of differently

functionalized 4H-3,1-benzothiazine (5e6) and 2H-1,4-benzothiazine (7) series as superior homologues

of 1. Biological evaluation demonstrated that amidine 4H-3,1-benzothiazine derivatives 5b-d can reduce

glutamate and LDH release in the oxygen/glucose deprivation and reperfusion model (OGD/R) applied to

brain slices with a higher potency than 1. Moreover the mentioned compounds signiﬁcantly reduce

glutamate- and 6-hydroxydopamine (6-OHDA)-induced cytotoxicity in neuroblastoma cells. In addition,

the same compounds limit ROS formation in both neuronal preparations. Finally, 5c proved effective in

inhibiting neuronal voltage-dependent Na^þand Ca^2þ-channels, showing a proﬁle comparable with that

of 1.

Received 28 September 2016

Received in revised form

8 November 2016

Accepted 26 November 2016

Available online 27 November 2016

Keywords:

Neuroprotective agents

Neurodegenerative diseases

Benzothiazine derivatives

SH-SY5Y cells

Neuronal voltage-dependent Na^þand Ca^2þ

currents

1. Introduction

modulate multiple pathways involved in the onset and progression

of the pathology may slow and deﬁnitely stop the evolution of the

Chronic neurodegenerative disorders, such as Alzheimer's dis-

ease (AD), Parkinson's disease (PD), Huntington's disease (HD), and

amyotrophic lateral sclerosis (ALS), as well as acute neurodegen-

erative diseases like brain ischemia, represent severe pathologies in

the developed world. The impact of these neurodegenerative dis-

eases is dramatically increased due to the society ageing with huge

human and economic costs [1]. Currently, the pathogenesis of the

mentioned disorders is not fully understood, and none of the

currently available treatments can stop their progression. In this

scenario, it is claimed that the administration of drugs able to

disease itself [2]. The aetiology of diverse neurodegenerative dis-

eases, however, is a complex process involving conjoint genetic and

cellular mechanisms, which dynamically interact to cause a com-

mon ﬁnal, neurodegenerative outcome [3]. In particular, several

experimental data strongly suggest that excitotoxic-mediated

neuronal damage plays a crucial role in the early pathogenesis of

chronic as well as acute neurodegenerative diseases [4]. Over-

activation of glutamate receptors impairs cellular calcium homeo-

stasis with consequent activation of neuronal nitric oxide synthase

(nNOS). Nitric oxide (NO) overproduction generates free radicals

and trigger programmed cell death. Simultaneous inhibition of key

signals involved in excitotoxic cell death will provide interesting

and efﬁcacious potential approaches for therapeutic intervention.

* Corresponding author.

E-mail address: maurizio.anzini@unisi.it (M. Anzini).

http://dx.doi.org/10.1016/j.ejmech.2016.1 1 .053

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

615

This can be attained either by means of a combination of drugs each

endowed with different mechanisms of action, or by means of

multi-target-direct ligands [5,6].

inhibiting properties of aminoguanidine [24] and

L

-NAME [25]

(Fig. 1). Thiourea derivatives 3 (Fig. 1) were synthesized since

several thioureas showed potent free radical scavengers activity,

preventing oxidative damage [26].

Since each step of the excitotoxic cascade could represent an

attractive drug target for the development of neuroprotective

agents beneﬁcial for the treatment of numerous chronic and acute

brain diseases, we developed new compounds endowed with

neuroprotective activity and particularly focused on those poten-

tially useful in the treatment of ALS. Our attention was ﬁrstly

attracted by riluzole (1, 2-amino-6-(triﬂuoromethoxy)benzothia-

zole) (Fig. 1), which has been shown to possess neuroprotective

effects in animal models of PD [7], HD [8] and cerebral ischemia [9].

Interestingly, 1 is a small molecule (MW ¼ 234.2) that in vitro

may elicit multiple molecular actions, among which those clinically

relevant are inhibition of voltage-gated sodium channels [10e13],

which can lead to reduced neurotransmitter release, non-

competitive inhibition of NMDA receptors [14,15], inhibition of

glutamate release [16], and enhanced astrocytic uptake of extra-

cellular glutamate [17].

Derivatives 2e4 were tested using an in vitro protocol of

ischemia/reperfusion injury and the results revealed that 2c and

3a-d meaningfully reduced neuronal injury. In particular, com-

pounds 3a-d were selected for evaluating their antioxidant prop-

erties. The results proved that the mentioned compounds were

endowed with a direct ROS scavenging activity. Compounds 3b and

3d underwent electrophysiological studies in order to assess their

potential activities on voltage-dependent Na^þand Ca^2þcurrents in

neurons from rat piriform cortex. Compound 3b inhibited the

transient Na^þcurrent at 50

[22].

mM, but to a much smaller extent than 1

Encouraged by our results, we decided to pursue the synthesis

of two different series of benzothiazines 5e7 (Fig. 2), as cyclic an-

alogues of benzothiazole.

Currently, 1 is the only drug that has proved able to modify the

progression of ALS [18e20] and the only one approved for the

treatment of this disease.

On these bases we designed and reported [21,22] the synthesis

and the biological evaluation of amidine and guanidine derivatives

2 and 4, respectively (Fig. 1), as 1-like compounds aiming to con-

jugate the neuroprotective effects of 1 with the neuroprotective

and anti-inﬂammatory activity of 1400W [23] and the NOS-

2. Results and discussion

2.1. Chemistry

The synthesis of 2-amino-4H-3,1-benzothiazines 15 and 16 has

been accomplished as reported in Scheme 1. The protection of the

amino group of p-triﬂuoromethoxyaniline 8 by means of pivaloyl

acid chloride in the presence of Et₃N gave pivaloyl amide 9 [27] that

Fig. 1. Reference compounds.

616

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

Fig. 2. Structures of target compounds 5e7.

was in turn formylated with t-BuLi in dry DMF to afford aldehyde

10. Reduction of this compound with NaBH₄in EtOH gave benzyl

alcohol 11 that, by treatment with concentrated HCl in dioxane, was

transformed into the corresponding chloromethyl derivative and

contemporary deprotected to yield 4-triﬂuoromethoxyaniline 12.

The condensation of compound 12 with thiourea in isopropyl

alcohol afforded the expected 2-amino-6-triﬂuoromethoxy-4H-3,1

benzothiazine 15. As for the synthesis of unsubstituted 2-amino-

4H-3,1-benzothiazine 16, 2-aminobenzyl alcohol 13 was used as

the starting material, and subjected to the same reactions used for

the o-triﬂuoromethoxy derivative 15. The transformation of benzyl

alcohol 13 into chloromethyl derivative 14 and successive cycliza-

tion with thiourea yielded compound 16 in quantitative yield [28].

o-Amino-S-cyanomethylthiophenol 18 used for the preparation

of 3-amino-7-triﬂuoromethoxy-2H-1,4-benzothiazine 19 was pre-

pared by hydrolysis of 1 followed by alkylation with chlor-

oacetonitrile in the presence of (Bu)₄N^þHSO₄^ꢀand 10 M NaOH.

Compound 18 was readily cyclized by reﬂuxing in ethanolic HCl

(5%) to give the desired compound 19. As for the synthesis of

unsubstituted 3-amino-2H-1,4-benzothiazine 21 the o-amino-

tiophenol 20 was used as the starting material and subjected to the

same alkylation and cyclization procedures as reported above to

afford the expected compound 21 [29] (Scheme 2).

in CH₃CN as described in literature [30] (but in better yields, 80%).

Compounds 15, 16, 19, and 21 were transformed into the cor-

responding amidines 5a-f, 7a-f and thiourea 6, respectively, ac-

cording to well-established methodologies (Scheme 4) [22,31].

2.2. Biological evaluation

The neuroprotective properties of the compounds reported in

Table 1 were ﬁrst investigated by means of rat brain slices, an

experimental model that preserves the exact cellular architecture

of an intact brain and that can be used for an appropriate and ac-

curate extrapolation of ﬁndings in terms of neuroprotection. Since

excitotoxicity is a key step of the ischemic cascade, which led to

neuronal loss, rat brain slices were exposed to ischemia-like con-

dition (oxygen glucose deprivation and reperfusion, OGD/R). The

amount of LDH and glutamate release from the tissue to the me-

dium was taken as an index directly related to tissue damage and

inversely related to neuroprotection. Compounds were added after

the injury since, from a clinical point of view, this is more relevant

than a treatment performed before the damage. Results demon-

strated that among the newly tested compounds 5b-d were the

most effective in reverting OGD/R-induced LDH and glutamate

release. As observed for 1 [22], however, their effects followed a U-

shaped, hormetic-like concentration-response curve (Fig. 3).

Compound 1, used to prepare compound 19, is quite expensive,

then we decided to prepare it in house according to an already pub-

lished procedure (Scheme 3) [30]. Brieﬂy, 1 was prepared via one-pot

procedure based on the condensation of 8 with 1 molar equivalent of

ammonium thiocyanate and benzyltrimethylammonium tribromide

For LDH release, the efﬁcacy window was 0.01e0.1

0.1e5 M for 5c and 5d, while the minimal effective concentration

(MEC) was 0.01 M and 0.1 M, respectively, i.e. values that differ by

one or two orders of magnitude from that of 1 (1 M). These

mM for 5b or

m

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

617

Scheme 3. Reagents: (i) NH₄SCN, PhCH₂N(CH₃)₃Br₃, CH₃CN.

Scheme 4. Reagents: (i) CH₃CONR₁R₂, POCl₃, dry toluene; (ii) R₁NCS, Et₃N, dry toluene.

R₁is deﬁned in Fig. 2.

glutamate release (~50e60%) and, at the same time, did not modify

that of LDH. This can be explained by considering the possibility

that 15 and 6 could hamper the steps of ischemic cascade that lead

to glutamate release into the extracellular space without affecting

those which lead to neuronal death. 7a and 7d exert poor neuro-

protection since they partially reverted the release of the endo-

cellular enzyme into the medium, leaving that of glutamate

unchanged (i.e. 7d) or only moderately reduced (i.e. 7a). Com-

pounds 7b,c and 5e, however, showed no activity.

To further characterize the neuroprotective effects exerted by

the most interesting compounds (5b-d) at the cellular level, the

human neuroblastoma cell line SH-SY5Y, widely employed in

studies related to neurotoxicity, oxidative stress and neurodegen-

erative diseases was used [32,33]. SH-SY5Y cells were incubated

Scheme 1. Reagents: (i) (CH₃)₃CCOCl, Et₃N, CH₂Cl₂; (ii) t-BuLi, dry THF, dry DMF; (iii)

NaBH₄, absolute EtOH; (iv) HCl 37%, dioxane; (v) HCl 37%, sealed vial; (vi) NH₂CSNH₂, i-

PrOH.

derivatives had a maximum effect of about 78% (5b) and concen-

trations higher than the efﬁcacy window were less active or inef-

fective in exerting neuroprotection. Regarding the ability of the

tested compounds to counteract OGD/R-induced glutamate release,

5c and 5d had a wider efﬁcacy window than 1, and they also

effectively reverted glutamate release at a concentration of 0.1

(5c) or 1 M (5d), respectively.

mM

m

5a, 7a, 7d, 15, and 6 were less active than 1 and the above-

mentioned compounds. In particular, 15 and 6 partially reverted

with 0.1, 1 and 10

to the cells the neurotoxins glutamate or 6-hydroxydopamine.

mM concentrations of 5b-d for 1 h before adding

Scheme 2. Reagents: (i) NaOH 10 M; (ii) ClCH₂CN, (Bu)₄N^þHSO₄^ꢀ, NaOH 10 M, CH₂Cl₂; (iii) HCl/EtOH 5%; (iv) ClCH₂CN, (Bu)₄N^þHSO₄^ꢀ, NaOH 25%, CH₂Cl₂.

618

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

Table 1

Effects of 1 and compounds 5e-f, 6, 7a-d and 15 on OGD and reperfusion-induced release of glutamate (GLU) and LDH in rat cortical brain slices.

Compounds

GLU

LDH

Efﬁcacy window^a

(m

M)

MEC^b(% SEM)^c

Efﬁcacy window (

m

M)

MEC (% SEM)

1 Riluzole^d

1.0e25

5.0e50

0.01e0.1

0.1e25.0

N.E.

0.01e1.0

0.001

0.1e100

N.E.

1 (100)

0.1e100

1

0.01e1.0

0.1e5.0

0.1e1.0

N.E.

0.1

N.E.

100

N.E.

10.0 (100.0 4.5)

0.1 (40.1 5.3)

0.01 (78.4 2.7)

0.1 (72.4 5.9)

0.1 (58.2 3.8)

N.E.

0.1 (31.0 2.1)

N.E.

100 (33.2 9.2)

N.E.

5a

5b

5c

5d

5e

5f

5 (43.6 5.7)

0.01 (72.7 10.3)

0.1 (100.0 4.4)

1 (90.5 3.6)

N.E.

0.01 (17.8 5.8)

0.001 (45.7 8.9)

0.1 (25.5 6.8)

N.E.

6

7a

7b

7c

7d

15

N.E.

0.001e0.1

N.E.

10

N.E.

10 (53.5 5.7)

N.E.

0.001 (69.1 6.8)

a

The efﬁcacy windows represent the interval of concentrations at which a signiﬁcant reduction of OGD-induced GLU and LDH release was observed.

MEC (Minimal Effective Concentration) is the mM concentration at which the highest reduction was observed.

The value between parentheses represents the % of reversion exerted at such concentration;100% was taken as the return to basal values (CTRL).N.E.: inactive.

Data from Ref. [22].

b

c

d

Fig. 3. Effects of compounds 5b, 5c and 5d on the release of glutamate and LDH induced by oxygen/glucose deprivation and reoxygenation (OGD/R) in rat cortical slices. Drugs were

added to reoxygenation buffer used after OGD. Data are mean SEM of at least four different experiments. Statistical analysis was performed by applying ANOVA followed by a post

hoc Dunnett test. ^ꢁꢁꢁp < 0.01 vs controls (CTRL). *p < 0.05, **p < 0.01, ***p < 0.001 vs OGD/R.

Riluzole, taken as reference drug, was tested as well. Results

demonstrated that all tested compounds exerted neuroprotection

against glutamate-induced injury (Fig. 4A); 10 mM or 0.01 mM were

non-neuronal cellular elements as well as neurotransmitter in-

teractions typical of a tissue context, is crucial for neuroprotection

and could hamper in some way the effects of riluzole but not those

of 5b-d.

Changes in SH-SY5Y cell cycle was also assessed. As illustrated in

Fig. 4B, the percent of sub G₀/G₁hypodiploid cells was signiﬁcantly

the most effective concentrations for 5b-d or riluzole, respectively,

since cells recovery that occur was the highest observed (~65e70%).

It should be noted that in this experimental model the concentra-

tions of 5b-d necessary for the neuroprotective effect were orders

of magnitude greater than that of riluzole. This however was not

the case of rat brain slices in which the novel derivatives-mediated

neuroprotection was observed at concentration close or even lower

than that of riluzole (see Table 1). This suggest that the presence of

increased after glutamate challenge (þ17.4

CTRL), while those in the S and G₂/M phase were decreased.

Interestingly, the pretreatment with selected compounds at 10

concentration (or 0.01 M for Riluzole) signiﬁcantly reduced the

percentage of cells in sub G₀/G₁, and at the same time, it increased

1.9%, p < 0.001 vs

m

M

m

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

619

Fig. 4. Effects of 1 h pretreatment with 5b, 5c, and 5d (0.1e10

Cell viability assessed by MTT assay. (B) Percentage of cells in the sub G₀/G₁(apoptotic) phase determined by ﬂow cytometry after propidium iodide staining. (C) DAPI staining

quantitative analysis. In panels B and C tested concentrations were 10 M (5b-5d) or 0.01 M (RIL). (D) Intracellular ROS production determined by means of the peroxide-sensitive

mM) or Riluzole (RIL, 0.001e0.01 mM) on glutamate (GLU)-induced cytotoxicity (50 mM for 24 h) in SH-SY5Y cells. (A)

m

ﬂuorescent probe DCFH-DA and expressed as percent inhibition of intracellular ROS produced by GLU. Values are reported as mean SEM of 4 ind_ꢁe_ꢁp_ꢁendent experiments (triplicate

or quadruplicate were performed in each experiment). Statistical evaluation was performed by means of ANOVA followed by Dunnet post test.

p < 0.001 vs CTRL. *p < 0.05,

**p < 0.01, ***p < 0.001 vs GLU.

or left unchanged that in G₀/G₁or in S phase.

treatment of SH-SY5Y cells with the above compounds and rilu-

zole elicited a concentration- and time-dependent decrease in ROS

production caused by glutamate. The maximal inhibitory effect was

SH-SY5Y morphological analysis was in agreement with previ-

ous data. The characteristic features of cell apoptosis induced by

glutamate (i.e. cell tendency to round-up and detach from the

culture plate, increased number of shrunken cell bodies with

broken neurites, grade 2e3 toxicity) were, in fact, reverted by the

selected compounds. This was further substantiated by data ob-

tained by DAPI staining. The percent of apoptotic cells, which

contained condensed and fragmented nuclear material, in fact, was

achieved at 10-mM for both 5c and 5d (24 h), at which concentra-

tion a reduction of about 20e40% in ROS formation was measured.

The neurotoxin 6-hydroxydopamine (6-OHDA), which is widely

used as a neuronal damage inductor in experimental models of PD,

has been reported to cause neuronal death by apoptosis, necrosis,

or a combination of the two [34]. The ability of compounds 5b-

d and riluzole to revert 6-OHDA-induced injury was thus also

dramatically increased after glutamate treatment (þ16.9

1.5%,

p < 0.01 vs CTRL) (Fig. 4C). By contrast, the percent of cells dis-

playing the ﬂuorescence of the dye was signiﬁcantly lower upon

the pre-treatment with selected compounds (9e10%, p < 0.01 vs

GLU, see also Figure S1 and S2 in the Supplementary Material).

Finally, the effects of these compounds per se on SH-SY5Y cells

tested in SH-SY5Y cells. As illustrated in Fig. 5A, 6-OHDA (55

mM for

2 h) signiﬁcantly decreased cell viability (47.6 6.2%, p < 0.01),

while a 1 h pretreatment of SH-SY5Y cells with all the four com-

pounds prevented 6-OHDA toxicity. The highest concentration used

(10 mM for 5b-d, 0.01 for riluzole) was the most active in exerting

viability were also tested. 24 h treatment with 10

m

M of 5b-

neuroprotection, by causing an almost complete recovery in cell

viability.

d proved to be basically devoid of cytotoxic effects since only a

slight reduction in cell viability (~10%, grade 0e1) was observed

(data not shown).

ROS strongly contributes to cell death caused by oxidative

glutamate toxicity and for this reason their intracellular formation

was assessed by DCFH-DA assay. As shown in Fig. 4D, the pre-

Furthermore, compounds 5b-d reduced apoptotic cell death

induced by 6-OHDA as determined by DAPI staining (Fig. 5B and

Figure S2). A high amount of DAPI-positive cells were found after

treatment with 6-OHDA, strongly suggesting the involvement of

apoptosis in the neurotoxin-mediated SH-SY5Y death. On the

620

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

Fig. 5. Effects of 1 h pretreatment with 5b, 5c, and 5d (0.1e10

SY5Y cells. (A) Cell viability, assessed by MTT assay, was expressed as percent of that obtained by treating cells with 6-OHDA vehicle (ascorbic acid, AA). (B) DAPI staining

quantitative analysis performed after treatment with 10 M 5b-5d or 0.01 M RIL. (C) Intracellular ROS production determined by means of the peroxide-sensitive ﬂuorescent probe

mM) or riluzole (RIL, 0.0001e0.1 mM) on 6-hydroxydopamine (6-OHDA)-induced cytotoxicity (55 mM for 2 h) in SH-

m

DCFH-DA and expressed as the percent of inhibition of intracellular ROS produced by 6-OHDA. Data are speciﬁed as mean SEM of 4 independent experiments (triplicate or

ꢁꢁꢁ

quadruplicate were performed in each experiment). Statistical evaluation was performed by means of ANOVA followed by Dunnet post test.

p < 0.001 vs CTRL. *p < 0.05,

**p < 0.01, ***p < 0.001 vs 6-OHDA.

contrary, compounds 5b-d (10

m

M) or riluzole (0.01

m

M) reduced by

could contribute to reduce the level of glutamate release, thus

partially justifying the effects of 1, 5c and 5d. Accordingly, the ac-

tivity of 5c and 5d on neuronal voltage-dependent Na^þand Ca^2þ

currents and the comparison of the results with those found for 1

has been performed.

about 20% the number of condensed or fragmented nuclei upon 6-

OHDA treatment. Finally, to examine whether ROS formation

mediated by 6-OHDA could be hampered by these compounds,

H2DCF-DA assay was performed. In the presence of 5b-d a con-

centration- and time-dependent reduction of intracellular ROS

formation caused by the neurotoxin was observed (Fig. 5C). Pre-

treatment of the cells with 5b-d, indeed, afforded signiﬁcant pro-

Patch-clamp recordings [36] of voltage-dependent Na^þcurrents

were accomplished employing neurons of rat piriform cortex layer

II in brain slices as an experimental model (see the Experimental

Section for further details). Differently from acutely dissociated

neurons (see below), this preparation allowed for optimal Na^þ

current stability, as required to evaluate the activity of the tested

compounds. The transient Na^þcurrent (I_NaT) was studied by means

of a current-voltage (I/V) protocol which consists of 19 ms depo-

larizing step pulses at ꢀ75 to 20 mV in 5 mV increments, starting

from a holding potential of ꢀ80 mV (not shown). During compound

application, I_NaTamplitude was also monitored by repetitively

commanding (once every 10 s) a single 19 ms step at ꢀ20 mV

(Fig. 6C and D). 5c or 5d was applied through the bath perfusion at

tection already at 0.1

all compounds at 10

m

M, while maximum effects were achieved for

M.

2.3. Effects of compounds 5c and 5d on neuronal voltage-

dependent Na^þand Ca^2þcurrents

As already discussed elsewhere [22], compound 1 inhibited

voltage-dependent Na^þcurrents by a mechanism in which stabi-

lized the inactivated state of Na^þchannels, representing an

important basis of its neuroprotective action. Moreover, in pe-

ripheral sensory neurons [35] and cortical neurons [22] 1 has also

been proven to exert inhibitory effects on high-voltage-activated

(HVA) Ca^2þchannels. The inhibition of voltage-dependent Ca^2þ

inﬂux under conditions of metabolic distress and depolarization

50

mM.

After a saturating effect on the I_NaTamplitude was reached, the

drug's effect was quantiﬁed by measuring I_NaTinhibition at a

voltage level equal to the I/V peak plus 20 mV (Fig. 6A and B), rather

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

621

Fig. 6. Effects of benzothiazine derivatives 5c and 5d on I_NaTamplitude. (A, B). Currents recorded in two representative neurons (cell E8722 in A; cell B8711 in B), at the test potential

M 5c (A) or 5d (B). Arrows indicate the starting time of the depolarizing pulse. (C, D). Time course of the

of V_pþ 20 mV, under control conditions and during application of 50

m

effects of benzothiazine compounds on I_NaTin two representative neurons (cell B8722 in A; cell B8710 in B). Drugs were applied, for the time interval indicated by horizontal bars,

through the general perfusion. (E) Average percentages of I_NaTpeak-amplitude inhibition induced by the benzothiazine compounds. n ¼ 9 (5c) and 5 (5d). (F) Average percentages of

I_NaTpeak amplitude inhibited by 5c at three different concentrations (5, 15 and 50

m

M). n ¼ 5, 4 and 9, respectively.

than at the peak itself, to minimize the consequences of space-

clamp artifacts [37]. 50 M 5c reduced I_NaTamplitude by 51.9

Fitting of experimental data with single Boltzmann functions

(Fig. 7C, continuous lines) revealed that, at 15 mM, 5c induced a

m

2.1% (n ¼ 9), whereas 5d had a much weaker effect (15.9 8.5%

negative shift of the inactivation function V on average by

½

inhibition, n ¼ 5) (Fig. 6E). These effects were compared with the

11.8 2.4 mV (from ꢀ58.4 1.6 mV to ꢀ70.4 4.3 mV, n ¼ 5): this

51.6% inhibition exerted by 1 at 50

mM, as reported elsewhere for

compares with a ꢀ15.5 0.4 mV shift induced by 15

mM of 1 under

the same experimental conditions [22]. The effects of both 5c and

5d were partially reversible (Fig. 6C and D). Since the effect of

the same experimental conditions [22]. The slope coefﬁcient of the

inactivation function, instead, was not signiﬁcantly modiﬁed (from

50

m

M 5d on I_NaTamplitude was markedly and signiﬁcantly smaller

5.7

0.2 mV to 6.3

0.8 mV; p ¼ 0.4, t-test for paired data),

than that of the same concentration of 1 (p ¼ 0.015), the activity of

the former drug on Na^þcurrents was not further characterized, and

the investigation was focused on the effects of 5c. The effects of this

compound on Na^þcurrents were examined also at two other

whereas the maximal current amplitude was reduced, on average,

by 24.2 7.4% (n ¼ 5): this reduction was statistically signiﬁcant

(p < 0.01).

Similarly to 1 [22], 5c also inhibited the persistent component of

the Na^þcurrent (I_NaP) much more pronouncedly than I_NaT. The

drug's effects on the I_NaPrecorded in two representative neurons

are illustrated in Fig. 8. Compound 5c reduced the peak amplitude

of the I_NaPelicited by a 2 s depolarizing ramp protocol by 70.7

concentrations of the drug, 5 and 15

mM. Fig. 6F shows the

concentration-dependence of the inhibition caused by 5c on I_NaT

amplitude. Based on these data, it can be inferred that the IC₅₀of

the inhibitory effect induced by 5c is very similar to that observed

for 1 [22].

The ability of 5c to modify the voltage dependence of I_NaT

inactivation was then characterized and compared with that of 1.

The currents recorded in a representative neuron in the absence

5.2% at the concentration of 5

(n ¼ 5), and by 91.8 3.6% at 50

mM (n ¼ 5), by 77.0 6.9% at 15

mM

mM (n ¼ 4). These percent inhi-

bition values were much, and signiﬁcantly, greater than those

observed for I_NaTamplitude (p < 0.005 for all three concentrations

examined; t-test for paired data). The inhibitory effects of 5c on I_NaP

amplitude were not signiﬁcantly different from those observed for

and in the presence of 15 mM of compound 5c in response to the

application of a steady-state inactivation protocol are illustrated in

Fig. 7B. Fig. 7C shows the steady-state inactivation plots obtained

from the same experiment: it can be seen that 5c induced a marked

shift of inactivation voltage dependence in the negative direction.

the same concentrations of 1 (p > 0.25 both at 15 mM and 50 mM; t-

test for unpaired data). Moreover, the selective inhibitory effect of

5c on I_NaP, as compared to I_NaT, was already evident during the

622

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

charge carrier, and Ba^2þcurrents (I_Bas) were recorded. High-

voltage-activated (HVA) currents were elicited by commanding

depolarizing step or ramp protocols starting from a conditioning

potential of ꢀ60 mV. 5c and 5d were applied to the recorded

neuron using a focal perfusion system [39]. The effects of com-

pounds 5c and 5d on HVA currents were monitored by repetitively

commanding (once every 7 s) a voltage-clamp protocol consisting

of a 40 ms depolarizing ramp from ꢀ60 to þ30 mV, which returned

an “instantaneous” I/V relationship for total HVA currents (see

Fig. 9CeE). This instantaneous I/V closely matched that obtained

with standard step I/V protocols (not shown, but see Ref. [22]).

Current amplitude was measured at the peak of the instantaneous

I/V.

Application of benzothiazine derivatives 5c and 5d via focal

perfusion caused an inhibitory effect on barium current (I_Ba) peak

amplitude with a biphasic time course (Fig. 9A). Indeed, after

activation of the drug's local perfusion an initial, fast reduction of

the current was observed, followed by a further, slower reduction

phase which required 3e4 min to reach a plateau level. As shown in

Fig. 9B, the average plot of the time course of inhibition could be

properly ﬁtted with a double exponential function, with time

constant values that for 5d were 23.4 ms and 128.2 ms. The effect of

both drugs was partly reversible after wash out (n ¼ 12) (Fig. 9A).

The concentration-dependence of the effect of benzothiazine

derivatives on I_Bamaximal amplitude is illustrated in Fig. 9DeG. In

both cases, the average concentration-dependence plots were best

ﬁtted by the sum of two Hill functions, with IC₅₀values of 5.4

mM

and 180.9 M for 5c; and 4.4 M and 110.1 M for 5d. These results

m

suggest that the action of benzothiazine derivatives on HVA

channels is due to their interaction with at least two distinct

binding sites. On the contrary, reference compound 1 showed a

single IC₅₀value of 34.2 mM along with a n_Hillvalue of 0.82 [22], thus

suggesting that its activity is due to the interaction with a single

binding site. Similarly to 1 [22] the inhibitory effect of 5c and 5d on

I_Baamplitude was not voltage-dependent, since the percentage of

inhibition was constant over the range of test potentials explored to

activate I_Bas (ꢀ50 to þ30 mV; not shown). This will be documented

and discussed in more depth elsewhere.

Overall, the above results indicate that compound 5c does exert

major inhibitory effects on both voltage-activated Na^þcurrents and

HVA Ca^2þcurrents over a concentration range superimposable with

that over which its in vitro neuroprotective effects were observed.

Compound 5d was likewise effective in inhibiting HVA Ca^2þcur-

rents, but much less so as far as voltage-dependent Na^þcurrents

are concerned.

Fig. 7. Compound 5c induces a prominent negative shift in the steady-state inactiva-

tion curve of I_NaT. (A) Voltage-clamp protocol applied to study the voltage dependence

of I_NaTsteady-state inactivation. (B) Currents recorded in a representative neuron (cell

A8731) in response to the protocol shown in An under control conditions and during

application of 15 mM 5c. The currents recorded at some signiﬁcant conditioning po-

2.4. Structure-activity relationships (SAR)

tential levels (V_cond) are indicated by arrows. (C) Plots of the voltage dependence of I_NaT

steady-state inactivation for the currents shown in panel B. Peak-current values

measured for the various V_conds have been normalized for the maximal value observed

in each condition, and plotted as a function of V_cond. Both plots have been ﬁtted with a

single Boltzmann function (continuous lines). Fitting parameters were:

Although in the absence of a clear mechanism of action of the

most active compounds, we have tried to extract from the biolog-

ical results some information about structure-activity relationships

(SAR) concerning the molecular modiﬁcations taken into consid-

eration. Two different series of benzothiazines (5, 6 and 7) have

been designed as superior homologues of 1. In particular, single

point changes have been made to benzothiazole scaffold with the

aim to understand which part can play a critical role.

V

¼ ꢀ60.2 mV, k ¼ 5.4 mV (control); V ¼ ꢀ71.5 mV, k ¼ 5.2 mV (5c).

½

application of relatively brief (19 ms) step depolarizations (Fig. 8C),

indicating that the effect of 5c on I_NaP, similarly to that of 1 [38], is a

kinetically fast one.

Data from Table 1 showed how only 4H-3,1-benzothiazines 5b-

d, and to a lesser extent compounds 5a, 7a, 7d, and 15, based on the

same N-C(S)-N skeleton of the parent compound, are able to elicit a

better neuroprotective activity as compared to 1 itself. On the other

hand, the inclusion of the N-C(C-S)-N skeleton in the cyclic system

of 2H-1,4-benzothiazine 6, yields a series of compounds that

proved inactive (for further details see Figure S3).

A summary of the average values of percent inhibition induced

by 5c and 5d, and by 1 for comparison, at the various concentra-

tions tested, on the transient and persistent components of the

voltage-dependent Na^þcurrent is provided by Table 2.

Voltage-dependent Ca^2þcurrents were evaluated in acutely

dissociated neurons from rat piriform cortex layer II [39], in which

they demonstrated to be highly stable and optimal clamp condi-

tions could be achieved. Ba^2þwas used instead of Ca^2þas the

The insertion of a methylene group between the aromatic ring

and the S atom in the 4H-3,1-benzothiazines 5b-d allows one to

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

623

Fig. 8. Compound 5c potently inhibits I_NaP. (A, B) Currents (A2, B2) recorded in response to the ramp protocol applied to activate I_NaP(A1, B1) in two representative neurons (cell

A8731 in A1; cell B8723 in B2) under control conditions and during application of 15 M or 50 M 5c. The currents shown are TTx-subtracted and are plotted as a function of

command potential. (C) The inhibition induced by 15

M of 5c on the transient and persistent components of the voltage-activated Na^þcurrent recorded in response to a 19 ms

m

depolarizing voltage pulse at 0 mV. The inset shows, over an expanded amplitude scale, a highlight of the persistent currents recorded during the late phase of the depolarizing

pulse. Note that even during such relatively short a depolarization 5c inhibited the persistent current component much more prominently than the early, transient component.

Same cell as in panel A. (D) Average effects of 5c at three different concentrations (5, 15 e 50

mM) on the peak amplitude of I_NaP(evoked by the slow ramp protocol). Bars represent

the average values of residual I_NaPamplitude observed in the presence of 5c, expressed as percent of control I_NaPamplitude. n ¼ 5 (5 mM), 4 (15

m

M) and 4 (50 M).

m

Table 2

(6) moieties with the aim to evaluate whether the amino group is

directly involved in the biological activity. Biological data from

Table 1 show that thiourea derivative (6) is completely inactive,

whereas amidine derivatives (5e7) proved active only in the

absence of OCF₃group. The simultaneous presence of OCF₃group

along with the disubstituted amidine moiety in position 2 of the

ring leads to inactive compounds due to a dramatic inﬂuence of the

electrostatic distribution of the molecular surface (data not shown).

Probably, triﬂuoromethoxy group as a long-range electron-with-

drawing substituent [40] is able to considerably decrease the ba-

sicity of amidine moiety causing a shift of pK_athat can modify the

binding afﬁnity properties of the molecules. Furthermore, by

means of the suitable alkyl substituents at the amidine nitrogen it is

Average values of the percent inhibition induced by compounds 5c and 5d on the

amplitude of the transient and persistent components of the voltage-dependent Na^þ

current.

Compounds

Concentration (

m

M)

Percent inhibition

I_NaT

I_NaP

1^a

15

50

5

15

50

39.4 7.1 (4)

51.6 5.6 (6)

26.4 7.2 (5)

36.0 10.0 (4)

51.9 2.1 (9)

15.9 8.5 (5)

87.3 4.3 (4)

94.8 9.0 (3)

70.7 5.2 (5)

77.0 6.9 (5)

91.8 3.6 (4)

n. a.

5c

5d

a

Data for 1 were taken from Ref. [22], are also given for comparison. Numbers of

observations are indicated between parentheses. n.a., not available.

possible to modulate the lipophilicity of the compounds,

a

parameter required for a good binding afﬁnity to the target protein.

Obviously, this seems not be true for 1 and 4H-3,1-benzothiazine 15

in which the absence of the lipophilic amidine moiety is fully or

partially compensated by OCF₃group leading to an active and a

moderately active compound, respectively.

Furthermore, in comparison with the inactive thiourea deriva-

tive 6, the presence of disubstituted amidine moiety in 2-position

of the most effective benzotiazines 5d seems to be crucial in

reverting OGD/R-induced LDH and glutamate release, with com-

pounds 5c and 5d showing a wider efﬁcacy window than 1. As a

whole, the transformation of the scaffold of 1 enlarging the ben-

zothiazole nucleus to have 4H-3,1-benzothiazines decorated with

the suitable lipophilic amidine moiety still leads to novel agents

endowed with neuroprotective properties.

obtain molecules less rigid without modifying the three-

dimensional arrangement of N-C(S)-N atoms, typical of the rilu-

zole system. For 2H-1,4-benzothiazine 7, instead, the methylene

insertion creates a more ﬂexible structure, suggesting that a rigid

skeleton constitutes the main pharmacophore requirement to

obtain novel active compounds able to protect against excitotoxic

insult.

Regarding the substituent on the aromatic ring, it can be infer-

red that the presence of triﬂuoromethoxy group in position 6 of the

ring does not seem to be essential for the neuroprotective activity:

in fact, unsubstituted 5b-d turned out to be the most active com-

pounds within the series.

As mentioned before, we introduced amidine (5e7) or thiourea

624

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

Fig. 9. Compounds 5c and 5d inhibit neuronal HVA Ca^2þcurrents. (A) Time course of I_Baamplitude changes in response to the application of 60

(cell C7517). Each data point represents the peak amplitude of a Ba^2þcurrent (I_Ba) evoked by the ramp protocol depicted in panel C. 5d was applied, for the time interval indicated by

the black horizontal bar, via focal perfusion. Note the reversibility of the drug effect after wash out. (B) Average plot of the time course of inhibition induced by 5d (60 M) on I_Ba

mM of 5d in a representative neuron

m

amplitude. In each cell, peak I_Baamplitude values have been used to calculate, for each data point, the quantity [1 þ (I_Baꢀ I_ss)/(I₀e I_ss)] ꢂ 100, where I_ssis the current level observed

after the development of a saturating effect of the drug, and I₀is current amplitude before drug application: this quantity represents therefore the total inhibition percentage

observed over time in that cell. The values thus obtained were then mediated among cells (n ¼ 7). The average plot has been ﬁtted with a double exponential function, with the

following ﬁtting parameters: A₁¼ ꢀ51.3, t₁¼ 23.4 ms, A₂¼ ꢀ41.3, t₂¼ 128.2 ms. (C) Ramp voltage-clamp protocol applied to evoke “instantaneous” I/V relationships for I_Bas. (D, E)

I_Bas recorded in two representative neurons (cell B7O22 in D; cell C7515 in C) in response to the protocol illustrated in panel C under control conditions and in the presence of

increasing concentrations of 5c (D) and 5d (E). (F, G) Concentration-dependence plots of percent inhibition induced by 5c (F) and 5d (G) on I_Baamplitude. Single data points

represent the average of values obtained from 4 (F) and 7 (G) cells. Data points have been ﬁtted with the sum of two Hill functions (continuous lines). Fitting parameters were:

A₁¼ 33.0%, IC_50-1¼ 5.4

m

M, n₁¼ 2.0, A₂¼ 67.0%, IC_50-2¼ 180.9

mM, n₂¼ 1.0 (5c); A₁¼ 22.7%, IC_50-1¼ 4.4

mM, n₁¼ 2.0, A₂¼ 77.3, IC_50-2¼ 110.1

mM, n₂¼ 1.8 (5d); IC₅₀¼ 34.2

mM,

n ¼ 0.82 (1 as reported in Ref. [22]).

As far as the activity of the novel compounds on voltage-

dependent Na^þchannels is concerned, compound 5c was the one

that showed the highest potency as an inhibitor of voltage-

activated Na^þcurrents, since it maintained a similar activity on

the transient and persistent current components (I_NaTand I_NaP) as 1.

Compound 5d, instead, has proven to be markedly less effective at

the same concentrations. Based on these results, it can be

concluded that the substitution of benzothiazole ring of 1 with the

4H-3,1-benzothiazine nucleus maintains the pharmacological ac-

tivity of 1 on voltage-gated Na^þchannels, provided that the

aliphatic residues of acetamidine group are not too extended.

3. Conclusions

Amidine and thiourea derivatives 5e7 were synthesized, and

their ability to counteract the excitotoxic cascade was evaluated. In

an in vitro protocol of ischemia/reperfusion injury, 4H-3,1-

benzothiazines 5b-d signiﬁcantly attenuated neuronal injury

showing to be the most effective derivatives in reverting OGD/R-

induced LDH and glutamate release in a concentration range be-

tween 0.1 and 25 mM. The same concentration-dependent neuro-

protective effects were also observed in the human neuroblastoma

cell line SH-SY5Y, commonly used to study neurotoxicity. Com-

pounds 5b-d counteracted apoptosis and ROS formation caused by

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

625

the treatment with glutamate or 6-hydroxydopamine, two exper-

imental models for excitotoxic- or oxidative stress-mediated injury.

Compound 5c proved to maintain basically the same activity of 1

sulfate, ﬁltered, and concentrated under reduced pressure. The

residue was puriﬁed by ﬂash-chromatography with ethyl acetate/

petroleum ether (1:1 v/v) as the eluent to give 16 as a light yellow

solid (1.7 g, yield 64%.). Re-crystallization from EtOH gave an

analytical sample melting at 132e135 ^ꢁC. ¹H NMR (CDCl₃): 3.90 (s,

2H), 5.09 (br s, 2H), 7.00e7.12 (m, 3H) 7.20e7.28 (m, 1H). MS (ESI):

m/z 165 (M þ H^þ). Due to its apparent instability compound 17 was

transformed into the correspondent oxalate which after a cautious

washing with dry diethyl ether gave a white solid melting at

178e182 ^ꢁC.

on the transient and persistent current components (I_NaTand I_NaP

)

of neuronal voltage-dependent Na^þcurrents. On the contrary

compound 5d was markedly less effective at the same concentra-

tions. Compound 5c was able to exert a pronounced inhibitory ef-

fect on glutamate release limiting neuronal death much more

effectively than previously reported benzothiazole compounds 2e4

[22]. This fact may be explained, at least in part, by the high potency

with which 5, with respect to the above-cited compounds, in-

terferes with the function of voltage-gated Na^þchannels. Differ-

ently from benzothiazole derivatives [22], 5c and 5d also proved

effective as a HVA Ca^2þ-channel inhibitors over a concentration

range at least partly compatible with that of their neuroprotective

effects. Indeed, the concentration-dependence plots of 5c and 5d

effects on HVA I_Baamplitude revealed the presence of an inhibitory

component characterized by a relatively high potency, with IC₅₀of

4.1.2. 3-Amino-2H-1,4-benzothiazine (21) [29]

A mixture of 2-aminothiophenol 20 (1.6 mL, 15 mmol) in NaOH

25% (p/v) (10 mL) and chloroacetonitrile (0.95 mL, 15 mmol) in

CH₂Cl₂(20 mL) was stirred at room temperature in the presence of

a

catalytic amount of tetrabutylammonium hydrogen sulfate

(0.51 g, 1.5 mmol). After 20 h, the organic layer was separated dried

over sodium sulfate, ﬁltered, and concentrated under reduced

pressure. The residue was dissolved in ethanolic hydrochloric acid

(5%) (20 mL) and reﬂuxed for 2 h. The solvent was removed under

reduced pressure and the resulting solid was dissolved in water,

washed with CHCl₃, made alkaline with concentrated NH₄OH

(pH ¼ 9.0) and extracted with CHCl₃. The organic layer was dried

over sodium sulfate, ﬁltered, and concentrated under reduced

pressure. The obtained residue was puriﬁed by ﬂash-

chromatography with ethyl acetate/Et₃N (8:2 v/v) as eluent to

obtain compound 21 as a light-yellow solid (0.80 g, yield 32%). Re-

crystallization from ethyl acetate gave an analytical sample melting

at 168e172 ^ꢁC. ¹H NMR (CDCl₃): 3.18 (s, 2H), 4.38 (br s, 2H),

6.89e7.25 (m, 4H). MS (ESI): m/z 165 (M þ H^þ). Due to its apparent

instability compound 21 was transformed into the correspondent

oxalate which after a cautious washing with dry diethyl ether gave

a white solid melting at 182e184 ^ꢁC.

4e5 mM (which is one order of magnitude lower than that of 1 [22])

and effective concentration levels very similar to those observed for

neuroprotective effects.

Our results suggest that 4H-3,1-benzothiazine derivatives

inhibit HVA Ca^2þchannels in a concentration-dependent manner

interacting with at least two binding sites (contrary to compound

1 that likely interacts with only one site [22]). Moreover, the fact

that the inhibitory effect of these compounds developed with a

biphasic time course indicates the existence of two distinct

mechanisms of action characterized by different time kinetics, one

of which relatively slow. It is therefore possible that 4H-3,1-

benzothiazine derivatives activate multiple mechanisms

converging on HVA Ca^2þchannels, thus inducing their inhibition.

One of these mechanisms (probably the one developing with the

slowest kinetics) could depend on the activation of G proteins, as

previously suggested for the action of 1 on Ca^2þchannels in

dorsal-root ganglion neurons [35].

4.1.3. N-[4-(Triﬂuoromethoxy)phenyl]-pivaloylamide (9)

To a solution of 4-(triﬂuoromethoxy)-aniline 8 (1.0 g, 5.6 mmol)

and Et₃N (0.71 mL, 5.1 mmol) in dry CH₂Cl₂(15 mL) cooled at 0 ^ꢁC,

pivaloyl chloride (0.73 mL, 5.9 mmol) was added dropwise. The

reaction mixture was stirred for 18 h at room temperature, poured

into ice-water and extracted with CH₂Cl₂(3 ꢂ 20 mL). The com-

bined organic extracts were dried over Na₂SO₄, ﬁltered and

concentrated under reduced pressure to give 9 as white solid (1.1 g,

yield 75%). Re-crystallization from n-hexane gave 9 as colorless

needles (m. p. 104e107 ^ꢁC). ¹H NMR (CDCl₃): 1.30 (s, 9H), 7.15 (d,

J ¼ 8.8, 2H), 7.33 (br s, 1H), 7.53 (d, J ¼ 9.2, 2H). MS (ESI): m/z 262

(M þ H^þ).

4. Experimental section

4.1. Chemistry

Melting points were determined in open capillaries on a Gal-

lenkamp apparatus and are uncorrected. Merck silica gel 60

(70e230 or 230e400 mesh) was used for column chromatography.

Merck TLC plates, silica gel 60 F₂₅₄were used for TLC. ¹H NMR

spectra were recorded with Bruker AC 200 spectrometer in the

indicated solvents (TMS as internal standard): the values of the

chemical shifts are expressed in ppm and the coupling constants (J)

in Hz. Mass spectra were recorded on either a ThermoFinnigan

LCQ-Deca or an Agilent 1100 L C/MSD. Purity of compounds 5a-f, 6,

and 7a-f, was assessed by RP-HPLC and was found to be higher than

95%. An Agilent 1100 Series system equipped with a Zorbax Eclipse

4.1.4. N-[2-Formyl-4-(triﬂuoromethoxy)phenyl]-pivaloylamide (10)

t-butyllithium (1.7 M in pentane) (8.1 mL,13.9 mmol) was added

dropwise to a solution of pivaloyl amide 9 (1.5 g, 5.7 mmol) in dry

THF (50 mL) cooled at ꢀ75 ^ꢁC. After 1 h, the mixture was treated

with DMF (0.44 mL, 5.7 mmol) and stirred at the same temperature

for additional 45 min, then allowed to warm up to room temper-

ature. After 20 min, the mixture was poured in ice-water. The

organic layer was washed with 4 N HCl and brine, dried over

Na₂SO₄, ﬁltered and concentrated under reduced pressure. The

residue was puriﬁed by ﬂash-chromatography with CH₂Cl₂/petro-

leum ether (1:1 v/v) as eluent to obtain compound 11 as a colorless

oil (1.2 g, yield 72%). ¹H NMR (CDCl₃): 1.34 (s, 9H), 7.43e7.50 (m,

2H), 8.86 (d, J ¼ 9.0, 1H), 9.90 (s, 1H), 11.30 (br s, 1H), MS (ESI): m/z

290 (M þ H^þ).

XDB-C8 (4.6 ꢂ 150 mm, 5

mm) column was used in the HPLC

analysis with methanol-water (80:20) as the mobile phase at a ﬂow

rate of 1.0 mL/min. UV detection was achieved at 254 nm.

4.1.1. 2-Amino-4H-3,1-benzothiazine (16) [28]

A mixture of 2-aminobenzylalcohol 13 (2.0 g, 16 mmol) in

concentrated HCl (10 mL) was heated at 100 ^ꢁC for 15 min in a

sealed vial. The resulting precipitate was ﬁltered and washed with

diethyl ether to give 14 as a white solid that was solubilized in

isopropanol (20 mL) and treated with thiourea (1.3 g, 17 mmol). The

mixture was reﬂuxed for 20 h, the solvent was removed under

reduced pressure and the residue diluted with H₂O, made alkaline

with 2 N NaOH (pH ¼ 10), and extracted with CH₂Cl₂. The organic

phase was washed with water to neutrality, dried over sodium

4.1.5. N-[2-Hydroxymethyl-4-(triﬂuoromethoxy)phenyl]-

pivaloylamide (11)

A solution of compound 10 (1.2 g, 4.1 mmol) in absolute EtOH

626

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

(20 mL) cooled at 0 ^ꢁC was treated with NaBH₄(0.19 g, 4.9 mmol).

The mixture was stirred for 15 min at 0 ^ꢁC and at room temperature

for additional 30 min. The solvent was concentrated under reduced

pressure and the residue was diluted with H₂O and extracted with

CHCl₃(3 ꢂ 20 mL). The combined organic extracts were dried over

Na₂SO₄, ﬁltered and concentrated under reduced pressure to give

11 as colorless solid (1.2 g, yield 90%). Re-crystallization from n-

hexane gave an analytical sample melting at 71e73 ^ꢁC. ¹H NMR

(CDCl₃): 1.30 (s, 9H), 2.24 (t, J ¼ 5.9, 1H), 4.68 (d, J ¼ 5.9, 2H), 7.03

(m,1H), 7.16 (m,1H), 8.15 (d, J ¼ 8.9,1H), 8.86 (br s,1H). MS (ESI): m/

z 315 (M þ Na^þ).

CHCl₃. The organic layer was washed with H₂O, dried over Na₂SO₄,

ﬁltered and concentrated under reduced pressure. The resulting

residue was puriﬁed by ﬂash-chromatography with the indicated

solvent as eluent to give the expected acetamidines (5a-f, 7a-f).

4.1.10. N'-(4H-3,1-benzothiazin-2-yl)-N-methylacetamidine (5a)

Title compound was obtained as a light yellow solid starting

from compound 16 and puriﬁed with ethyl acetate as eluent (yield

40%). ¹H NMR (CDCl₃): 2.16 (s, 3H), 3.04 (s, 6H), 3.94 (s, 2H),

7.00e7.08 (m, 3H), 7.17e7.024 (m, 1H), 11.90 (br s, 1H). MS (ESI): m/z

220 (M þ H^þ). Due to its instability compound 5a was transformed

into the correspondent oxalate which after a cautious washing with

dry diethyl ether gave a solid melting at 110e113 ^ꢁC dec.

4.1.6. 2-Amino-6-triﬂuoromethoxy-4H-3,1-benzothiazine (15)

To a solution of compound 11 (0.35 g, 1.2 mmol) in dioxane

(30 mL) cooled at 0 ^ꢁC, HCl 37% (4 mL) was cautiously added. The

mixture was heated at 80 ^ꢁC for 4 h, and then diluted with iso-

propanol (20 mL) and added of thiourea (0.14 g, 1.8 mmol). The

resulting reaction mixture was reﬂuxed for 20 h and then

concentrated under reduced pressure. The residue was diluted with

H₂O, made alkaline with 2 N NaOH (pH ¼ 9) and extracted with

CH₂Cl₂(3 ꢂ 15 mL). The combined organic extracts were dried over

Na₂SO₄, ﬁltered and concentrated under reduced pressure. The

resulting residue was puriﬁed by ﬂash-chromatography with ethyl

acetate/petroleum ether (1:1 v/v) as eluent to give 15 as light yel-

low (0.25 g, yield 84%). Re-crystallization from EtOH gave an

analytical sample melting at 115e118 ^ꢁC. ¹H NMR (CDCl₃): 3.87 (s,

2H), 5.16 (br s, 2H), 6.97e7.11 (m, 3H). MS (ESI): m/z 249 (M þ H^þ).

4.1.11. N'-(4H-3,1-benzothiazin-2-yl)-N,N-dimethylacetamidine

(5b)

Title compound was obtained as a light yellow oil starting from

compound 16 and was puriﬁed with ethyl acetate as eluent (yield

63%). ¹H NMR (CDCl₃): 2.20 (s, 3H), 3.06 (s, 3H), 3.99 (s, 2H),

7.06e7.12 (m, 2H), 7.16e7.27 (m, 2H). MS (ESI): m/z 234 (M þ H^þ).

Due to its instability compound 5b was transformed into the

correspondent oxalate which after a careful washing with dry

diethyl ether gave a solid melting at 154e157 ^ꢁC dec.

4.1.12. N'-(4H-3,1-benzothiazin-2-yl)-N,N-diethylacetamidine (5c)

Title compound was obtained as a yellow oil starting from

compound 16 (1.2 mmol) and puriﬁed with petroleum ether-ethyl

acetate (1:1 v/v) as eluent (yield 38%). ¹H NMR (CDCl₃): 1.18 (t,

J ¼ 7.2, 6H), 2.19 (s, 3H), 3.42 (br s, 4H), 3.97 (s, 2H), 7.01e7.07 (m,

2H), 7.14e7.27 (m, 2H). MS (ESI): m/z 262 (M þ H^þ).

4.1.7. 2-[2-Amino-5-(triﬂuoromethoxy)phenylthio]-acetonitrile

(18)

A suspension of 2-amino-6-triﬂuoromethoxybenzothiazole 1

(1.8 g, 7.7 mmol) in 10 N NaOH (30 mL) was reﬂuxed under a slow

stream of nitrogen until the suspension turned to a clear solution.

Then, a mixture of chloroacetonitrile (0.48 mL, 7.7 mmol) in CH₂Cl₂

(50 mL) and tetrabutylammonium hydrogen sulfate (0.26 g,

0.77 mmol) were added. The reaction mixture was stirred for

18 h at room temperature, the organic layer was then separated,

washed with water, dried over Na₂SO₄and ﬁltered. The solvent was

removed under reduced pressure and the residue was puriﬁed by

ﬂash-chromatography with petroleum ether/ethyl acetate (1:1 v/v)

as eluent to give 18 as a brown oil (1.1 g, yield 58%). ¹H NMR (CDCl₃):

3.45 (s, 2H), 4.44 (br s, 2H), 6.74 (d, J ¼ 8.8,1H), 7.10 (m,1H), 7.39 (m,

1H). MS (ESI): m/z 249 (M þ H^þ).

4.1.13. N'-(4H-3,1-benzothiazin-2-yl)-N,N-dipropylacetamidine

(5d)

Title compound was obtained as a light yellow oil starting from

compound 16 (1.2 mmol), and puriﬁed with petroleum ether-ethyl

acetate (1:1 v/v) as eluent (yield 54%). ¹H NMR (CDCl₃): 0.91 (t,

J ¼ 7.4, 6H), 1.61 (m, 4H), 2.19 (s, 3H), 3.29 (br s, 4H), 3.98 (s, 2H),

7.06e7.08 (m, 2H), 7.16e7.25 (m, 2H). MS (ESI): m/z 290 (M þ H^þ).

4.1.14. N,N-diethyl-N'-[6-(triﬂuoromethoxy)-4H-3,1-benzothiazin-

2-yl]acetamidine (5e)

Title compound was obtained as a light yellow oil starting from

compound 15 (1.2 mmol) and puriﬁed with petroleum ether-ethyl

acetate (1:1 v/v) as eluent (yield 65%). ¹H NMR (CDCl₃): 1.24 (t,

J ¼ 7.3, 6H), 2.19 (s, 3H), 3.39 (br s, 4H), 3.95 (s, 2H), 6.95 (s, 1H),

7.04e7.18 (m, 2H). MS (ESI): m/z 346 (M þ H^þ).

4.1.8. 3-Amino-7-triﬂuoromethoxy-2H-1,4-benzothiazine (19)

A solution of 18 (0.70 g, 2.82 mmol) in EtOH with HCl 5% (20 mL)

was heated under reﬂux for 2 h. The solvent was concentrated

under reduced pressure and the residue diluted with H₂O. The

aqueous layer was washed with CHCl₃, made alkaline with

concentrated NH₄OH (pH ¼ 9) and the resulting precipitate was

extracted with CHCl₃. The organic layer was dried over Na₂SO₄,

ﬁltered and concentrated under reduced pressure to give 19 as a

light yellow solid (0.45 g, yield 64%, m.p. 94e96 ^ꢁC). ¹H NMR

(CDCl₃): 3.14 (s, 2H), 5.04 (br s, 2H), 6.92e7.10 (m, 3H). MS (ESI): m/z

249 (M þ H^þ).

4.1.15. N,N-dipropyl-N'-[6-(triﬂuoromethoxy)-4H-3,1-

benzothiazin-2-yl]acetamidine (5f)

Title compound was obtained as a light yellow oil starting from

compound 15 (1.2 mmol) and puriﬁed with petroleum ether-ethyl

acetate (65:35 v/v) as eluent (yield 58%). ¹H NMR (CDCl₃): 0.89 (t,

J ¼ 7.3, 6H), 1.61 (m, 4H), 2.13 (s, 3H), 3.30 (br d, 4H), 3.93 (s, 2H),

6.93 (m, 1H), 7.04 (d, J ¼ 9.6, 1H), 7.15 (d, J ¼ 8.7, 1H). MS (ESI): m/z

374 (M þ H^þ).

4.1.9. General procedure for the synthesis of acetamidines (5a-f, 7a-

f)

4.1.16. N'-(2H-1,4-benzothiazin-3-yl)-N-methylacetamidine (7a)

Title compound was obtained as an orange-yellow solid starting

from compound 21 and puriﬁed with ethyl acetate-Et₃N (9:1 v/v) as

eluent (yield 28%). ¹H NMR (CDCl₃): 2.16 (s, 3H), 3.05 (s, 3H), 3.29 (s,

2H), 6.92e7.27 (m, 4H), 12.50 (br s, 1H). MS (ESI): m/z 220 (M þ H^þ).

Due to its instability compound 7a was transformed into the

correspondent oxalate which after several washings with dry

diethyl ether gave a solid melting at 164e166 ^ꢁC dec.

To a solution of POCl₃(4.9 mmol) in dry toluene (20 mL), cooled

at 0 ^ꢁC, the suitable acetamide (2.7 mmol) was added. After stirring

under argon for 30 min at room temperature 2-amino- or 3-amino-

benzothiazine (15, 16, 19 or 21) (2.4 mmol) was added. The reaction

mixture was reﬂuxed for 3 h, cooled at room temperature and

poured into ice-water. Then, the mixture was made alkaline with a

solution of NaOH 2 N in water up to pH ¼ 9 and extracted with

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

627

4.1.17. N'-(2H-1,4-benzothiazin-3-yl)-N,N-dimethylacetamidine

(7b)

Title compound was obtained as an orange oil starting from

and used for assays during exponential phase of growth.

4.2.2. Solution of drugs

compound 21 and puriﬁed with ethyl acetate-Et₃N (95:5 v/v) as

eluent (yield 26%). ¹H NMR (CDCl₃): 2.24 (s, 3H), 3.05 (s, 6H), 3.19 (s,

2H), 6.90e6.97 (m, 1H), 7.07e7.27 (m, 3H). MS (ESI): m/z 234

(M þ H^þ). Due to its instability compound 7b was transformed into

the correspondent oxalate which after several washings with dry

diethyl ether gave a solid melting at 120e122 ^ꢁC dec.

Glutamate (GLU) solutions was prepared as already reported

[41,42], while 6-OHDA was prepared immediately before use by

dissolving the powder in 5 mM ascorbic acid solution previously

gassed with nitrogen for 30 min. Stock solution of tested com-

pounds were prepared in DMSO (1 ꢂ 10^ꢀ1M or 1 ꢂ 10^ꢀ2M) and

diluted to the desired ﬁnal concentration with PBS immediately

before use. Final DMSO concentration in the samples was always

lower than 0.01% and it did not affect investigated parameters.

4.1.18. N'-(2H-1,4-benzothiazin-3-yl)-N,N-diethylacetamidine (7c)

Title compound was obtained as a green oil starting from

compound 21 and puriﬁed with ethyl acetate as eluent (yield 16%).

¹H NMR (CDCl₃): 1.18 (t, J ¼ 7.9, 6H), 2.26 (s, 3H), 3.18 (s, 2H), 3.44

(br s, 4H), 6.89e6.97 (m, 1H), 7.10e7.27 (m, 3H). MS (ESI): m/z 262

(M þ H^þ).

4.2.3. Glutamate- or 6-hydroxydopamine-induced injury in SH-

SY5Y cells

SH-SY5Y cells were seeded into 96-well plates (8 ꢂ 10⁴cells/mL,

ﬁnal volume 200 mL) and allowed to attach for 24 h. Afterward cells

were pre-incubated with different concentration of tested com-

pounds for 1 h and then GLU (50 mM for 24 h) or 6-OHDA (55 mM

4.1.19. N'-(2H-1,4-benzothiazin-3-yl)-N,N-dipropylacetamidine

(7d)

for 2 h) were added to the wells. These treatments were selected in

preliminary experiments as they caused about 50% cell death [42].

After the injury, cell viability and apoptosis assays were performed

as detailed below. Finally, the untreated group represented the

control group (CTRL) which was treated with medium (GLU) or

Title compound was obtained as a brown oil starting from

compound 21 and puriﬁed with ethyl acetate as eluent (yield 9%).

¹H NMR (CDCl₃): 0.91 (t, J ¼ 7.4, 6H), 1.63 (m, 4H), 2.26 (s, 3H),

3.23e3.29 (m, 6H), 6.89e6.97 (m, 1H), 7.07e7.27 (m, 3H). MS (ESI):

m/z 290 (M þ H^þ).

medium containing 5 mM ascorbic acid (6-OHDA).

4.1.20. N,N-diethyl-N'-[7-(triﬂuoromethoxy)-2H-1,4-benzothiazin-

3-yl]acetamidine (7e)

4.3. Cell viability and morphological assay

Title compound was obtained as a yellow oil starting from

compound 19 and puriﬁed with ethyl acetate as eluent (yield 34%).

¹H NMR (CDCl₃): 1.19 (t, J ¼ 7.1, 6H), 2.27 (s, 3H), 3.18 (s, 2H),

3.25e3.47 (m, 4H), 6.96 (m, 1H) 7.11e7.20 (m, 2H). MS (ESI): m/z

346 (M þ H^þ).

3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium

bro-

mide (MTT) standard assay was used to assess cell viability [42]. In

order to monitor changes in cell morphology caused by the treat-

ment with neurotoxins, SH-SY5Y cells were observed under a

phase-contrast light microscope and photographs were taken.

The results were evaluated using the grade scale, described in

USP 28 (United States Pharmacopeia edition 2005) (grades 0e4) for

assessment of the cytotoxic potential of tested materials, as follows:

grade 0 - none reactivity (discrete intracytoplasmic granules, no

cell lysis); grade 1 - slight reactivity (no more than 20% of the cells

are round, loosely attached and without intracytoplasmic granules;

occasional lysed cells are present); grade 2 - mild reactivity (no

more than 50% of the cells are round and devoid of intracytoplasmic

granules, no extensive cell lysis and empty areas between cells);

grade 3 - moderate (up to 70% of cells are rounded or lysed); grade 4

- severe (nearly complete destruction of the cells).

4.1.21. N,N-dipropyl-N'-[7-(triﬂuoromethoxy)-2H-1,4-

benzothiazin-3-yl]acetamidine (7f)

Title compound was obtained as an orange oil starting from

compound 19 (1.2 mmol) and puriﬁed with petroleum-ethyl ace-

tate (1:1 v/v) as eluent (yield 62%). ¹H NMR (CDCl₃): 0.91 (t, J ¼ 7.3,

6H), 1.63 (m, 4H), 2.25 (s, 3H), 3.17e3.48 (m, 6H), 6.95 (d, J ¼ 8.9,

1H), 7.11e7.20 (m, 2H). MS (ESI): m/z 374 (M þ H^þ).

4.1.22. 1-(4H-3,1-benzothiazin-2-yl)-3-propylthiourea (6)

To a solution of compound 16 (0.60 mmol) in dry toluene

(20 mL) and Et₃N (0.10 mmol) stirred under argon atmosphere, the

propyl isothiocyanate (0.84 mmol) was added dropwise. The re-

action mixture was heated under reﬂux for 5 h and then cooled at

room temperature. The solvent was removed under reduced pres-

sure and the residue diluted with CH₂Cl₂. The resulting mixture was

washed with H₂O, and the organic layer was dried over Na₂SO₄,

ﬁltered and concentrated under reduced pressure. The obtained

residue was puriﬁed by washings with diethyl ether to obtain the

compound 6 as a light yellow solid (yield 40%). Re-crystallization

from ethyl acetate gave an analytical sample melting at

180e184 ^ꢁC. ¹H NMR (DMSO-d₆): 0.95 (m, 3H), 1.64 (m, 2H), 3.53

(m, 2H), 4.02 (s, 2H), 7.05e7.27 (m, 4H),10.78 (s,1H),11.99 (br s,1H).

MS (ESI): m/z 266 (M þ H^þ).

4.4. Apoptosis assays

Analysis of cell cycle and sub G₀/G₁population was used to

check for cell apoptosis by means of ﬂow cytometry. SH-SY5Y

(2.5 ꢂ 10⁵cells/mL, ﬁnal volume of 2 mL) were treated with

selected compounds and then subjected to GLU or 6-OHDA treat-

ment. After the toxic damage, cells were ﬁxed with 70% ethanol

(30 min at 4 ^ꢁC) and then kept at ꢀ20 ^ꢁC for at least 24 h.

DNA was extracted by using a FACScan ﬂow cytometer through a

563e607 nm band-pass ﬁlter (BD Biosciences, San Jose, CA, USA). At

least 1 ꢂ 10⁴cells sample were acquired and the percentage of

apoptotic cell (sub G₀/G₁peak) was quantiﬁed by using Cell Quest

software (BD Biosciences, San Jose, CA, USA).

4.2. Neuroprotection experiments

4.5. Intracellular ROS content

4.2.1. Cell cultures

Human SH-SY5Y neuroblastoma cells were maintained in RPMI

1640 medium supplemented with 10% FBS, penicillin (100 U/ml),

To detect the production of ROS, 2⁰, 7⁰-dichloroﬂuorescein

diacetate (DCFH-DA)-based assay was used. Before any treatment,

streptomycin (100 mg/mL), and

L

-glutamine (2 mM), in a humidi-

SH-SY5Y cells were loaded with the ﬂuorescent probe (5 mM,

30 min) and then softly washed in PBS. Fluorescence was measured

at 1 h or 24 h (wavelengths 485 nm excitation, and 535 nm

ﬁed atmosphere of 95% air and 5% CO2 at 37 ^ꢁC. Cells were sub-

cultured when they were 70e80% conﬂuent, fed twice each week

628

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

(i) Experiments on acute slices. 350 mm thick coronal sections of

emission). Antioxidant activity was expressed as the percent of

inhibition of intracellular ROS produced by GLU or 6-OHDA treat-

ment [42].

the anterior piriform cortex were cut under hypothermic

conditions. During recordings, slices were perfused with an

extracellular solution suitable for the isolation of Na^þcur-

rents and containing: 100 mM NaCl, 26 mM NaHCO₃,

19.5 mM tetraethylammonium chloride (TEA-Cl), 3 mM KCl,

2 mM MgCl₂, 2 mM CaCl₂, 2 mM BaCl₂, 0.5 mM CdCl₂, 4 mM

4.6. DAPI staining

DAPI (4⁰,6-diamidino-2-phenylindole) staining kit (Life Tecnol-

ogies Italia, Monza, Italy) was used to check for cells with nuclei

with fragmented and condensed, scored as apoptotic. Manufac-

turer's protocol, with minor modiﬁcations already described was

used [42].

4-aminopyridine (4-AP), and 11 mM D-glucose (pH 7.4 by

saturation with a 95% O₂/5% CO₂mixture). Patch pipettes

were ﬁlled with an intracellular solution containing 104 mM

CsF, 50 mM TEA-Cl, 2 mM MgCl₂, 10 mM N-(2-hydroxyethyl)

piperazine-N0-2-ethanesulfonic acid (HEPES), 10 mM

ethylene glycol bis(b

-aminoethyl ether) N,N,N⁰,N⁰-tetraacetic

4.7. Rat brain cortical slices

acid (EGTA), 2 mM adenosine 5⁰-triphosphate (ATP)-Na₂, and

0.2 mM guanosine 5⁰-triphosphate (GTP)-Na (pH adjusted to

7.2 with CsOH).

All animal care and experimental protocols were approved by

the Italian Department of Health (813/2015-PR) and conformed the

EU Guidelines for the Care and the Use of Laboratory Animals

(European Union Directive 2010/63/EU). The experimental pro-

tocols used were already reported [43,44].

The general holding potential of voltage-clamp recordings

was ꢀ80 mV. Current signals were low-pass ﬁltered and digitized at

cutoff and sampling frequencies of 5 and 50 kHz, respectively.

Currents were always online leak subtracted via a P/4 routine.

Tetrodotoxin (TTx) (Alomone Laboratories, Jerusalem, Israel) was

applied in the bath with the superfusing solution at the end of all

Brieﬂy, cortical slices (400 mm) were prepared from Wistar rats

(male, 250e300 g weight, Charles River Italia, Calco, Italy), and

incubated in oxygenated artiﬁcial cerebrospinal ﬂuid (ACSF,

composition in mM 120 NaCl, 2.5 KCl, 1.3 MgCl₂, 1.0 NaH₂PO₄, 1.5

CaCl₂, 26 NaHCO₃, 11 glucose, bubbled with 95% O₂-5% CO₂, ﬁnal

pH 7.4, osmolality 285e290 mOsmol) for 1 h at room temperature.

Afterward slices were placed in ﬂasks containing 2 mL ACSF

and equilibrated for other 30 min at 37 ^ꢁC. Oxygen glucose

deprivation (OGD) was achieved by incubating slices into

ischemic-ACSF (i.e. glucose-free ACSF added with 11 mM sucrose,

continuously bubbled with a 95% N₂/5% CO₂gas mixture) for

30 min. After the OGD phase, reoxygenation was reproduced by

replacing the ischemic-ACSF with fresh, oxygenated ACSF for an

additional 90 min period. Neuroprotection exerted by tested

compounds was investigated by adding them to the ACSF used

during reoxygenation [43].

recordings, and currents recorded in the presence of 1 mM TTx were

always subtracted from those recorded under control conditions

and in the presence of drugs to abolish residual, unsubtracted

capacitive and/or leakage currents.

(ii) Experiments on acutely dissociated neurons. A mechanical

and enzymatic dissociation procedure described previously

was applied [39]. After seeding in the recording chamber,

cells were perfused with an oxygenated extracellular solu-

tion suitable for isolating Ba^2þcurrents conducted through

Ca^2þchannels, containing: 88 mM choline-Cl, 40 mM TEA-Cl,

3 mM KCl, 2 mM MgCl₂, 5 mM BaCl₂, 3 mM CsCl, 10 mM

HEPES, 5 mM 4-AP, and 25 mM D-glucose (pH 7.4 with HCl).

Patch pipettes were ﬁlled with an intracellular solution

containing 78 mM Cs methanesulfonate, 40 mM TEA-Cl,

10 mM HEPES, 10 mM EGTA, 20 mM phosphocreatine di-

Tris salt, 2 mM ATP-Na₂, and 20 units/mL creatine phos-

phokinase (pH adjusted to 7.2 with CsOH). Tight seals

4.8. Assessment of brain slices injury

Brain slices injury was assessed by measuring the amount of

both glutamate and LDH released into the ACSF during the entire

reperfusion period as reported by Contarese et al. [43].

(>5 GU) and the whole-cell conﬁguration were obtained

according to the standard technique [36]. All recording

conditions were as previously described [39]. The general

holding potential of voltage-clamp recordings was ꢀ70 mV.

Current signals were ﬁltered at 5 kHz and digitized at 50 kHz.

(iii) Drug Application. Concentrated (30 mM) stock solutions of 1

and compounds 5c and 5d were prepared in DMSO, divided

into small aliquots, and stored at ꢀ20 ^ꢁC. The aliquots were

then diluted to the ﬁnal concentrations in one of the extra-

cellular solutions described above. Preliminary control ex-

periments indicated that DMSO had no signiﬁcant effects on

Ca^2þcurrents when applied at the same concentrations used

to dissolve drugs in aqueous solution (0.02e0.67%, v/v). In

experiments for recording Na^þcurrents in slices, the drug-

containing extracellular solution was applied through the

bath perfusion. In experiments for recording Ca^2þcurrents in

acutely dissociated neurons, drugs were applied through a

local perfusion system described elsewhere [22].

4.9. Data analysis

Data are expressed as mean SEM and were collected from at

least 3 independent experiments, each run in quadruplicate.

ANOVA (ordinary or repeated measures followed by Dunnett post

test, GraphPad Prism version 5.04, GraphPad Software Inc., San

Diego, CA, USA) was used to assess statistical signiﬁcance at the

signiﬁcance level p < 0.05.

4.10. Patch-clamp experiments

Whole-cell, patch-clamp recordings were performed in layer II

pyramidal neurons of rat piriform cortex (PC), either in slices or

after acute dissociation. All the procedures followed for obtaining

piriform-cortex slices and acutely-dissociated neurons and for

carrying out whole-cell recordings were exactly the same as

described elsewhere (see Ref. [22]), and below they will be only

brieﬂy summarized. Young (P15eP22) Wistar rats of either sex

were used, following a protocol that conformed with the rules

established by the University of Pavia for the use of animals in

experimental studies, in compliance with the guidelines of the

Italian Ministry of Health, the national laws on animal research (d.l.

116/92), and the EU guidelines on animal research (N. 86/609/CEE).

(iv) Data Analysis. Whole-cell current signals were analyzed us-

ing Clampﬁt from pClamp version 8.2 (Axon Instruments,

Union City, CA). Na^þcurrents were normally reﬁltered off-

line at 3.5 kHz. Current amplitude was measured at the

peak of each tracing. Average values were expressed as

arithmetic means

SEM. Statistical signiﬁcance was

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

629

[4] A. Mehta, M. Prabhakar, P. Kumar, R. Deshmukh, P.L. Sharma, Excitotoxicity:

bridge to various triggers in neurodegenerative disorders, Eur. J. Pharmacol.

698 (2013) 6e18.

evaluated by applying the two-tail Student's t-test for paired

or unpaired data, as appropriate.

[5] S. Butini, K. Nikolic, S. Kassel, H. Bruckmann, S. Filipic, D. Agbaba, S. Gemma,

S. Brogi, M. Brindisi, G. Campiani, H. Stark, Polypharmacology of dopamine

receptor ligands, Prog. Neurobiol. 142 (2016) 68e103.

Conﬂict of interest

[6] L. Ismaili, B. Refouvelet, M. Benchekroun, S. Brogi, M. Brindisi, S. Gemma,

G. Campiani, S. Filipic, D. Agbaba, G. Esteban, M. Unzeta, K. Nikolic, S. Butini,

J. Marco-Contelles, Multitarget compounds bearing tacrine- and donepezil-

like structural and functional motifs for the potential treatment of Alz-

heimer's disease, Prog. Neurobiol. (2016), http://dx.doi.org/10.1016/

j.pneurobio.2015.1012.1003.

The authors declare no competing ﬁnancial interest.

Acknowledgements

[7] E. Fernandez-Espejo, Pathogenesis of Parkinson's disease: prospects of neu-

roprotective and restorative therapies, Mol. Neurobiol. 29 (2004) 15e30.

[8] R.M. Bonelli, G.K. Wenning, H.P. Kapfhammer, Huntington's disease: present

treatments and future therapeutic modalities, Int. Clin. Psychopharmacol. 19

(2004) 51e62.

[9] O. Ates, S.R. Cayli, I. Gurses, A.B. Karabulut, N. Yucel, A. Kocak, C.O. Cakir,

S. Yologlu, Do sodium channel blockers have neuroprotective effect after

onset of ischemic insult? Neurol. Res. 29 (2007) 317e323.

[10] E. Benoit, D. Escande, Riluzole speci ﬁ cally blocks inactivated Na channels in

myelinated nerve ﬁ bre, P ﬂ ugers Arch. 419 (1991) 603e609.

[11] T. Hebert, P. Drapeau, L. Pradier, R.J. Dunn, Block of the rat brain IIA sodium

channel alpha subunit by the neuroprotective drug riluzole, Mol. Pharmacol.

45 (1994) 1055e1060.

[12] J.H. Song, C.S. Huang, K. Nagata, J.Z. Yeh, T. Narahashi, Differential action of

riluzole on tetrodotoxin-sensitive and tetrodotoxin-resistant sodium chan-

nels, J. Pharmacol. Exp. Ther. 282 (1997) 707e714.

[13] C. Zona, A. Siniscalchi, N.B. Mercuri, G. Bernardi, Riluzole interacts with

voltage-activated sodium and potassium currents in cultured rat cortical

neurons, Neuroscience 85 (1998) 931e938.

[14] M.W. Debono, J. Le Guern, T. Canton, A. Doble, L. Pradier, Inhibition by riluzole

of electrophysiological responses mediated by rat kainate and NMDA re-

ceptors expressed in Xenopus oocytes, Eur. J. Pharmacol. 235 (1993) 283e289.

[15] A. Doble, The pharmacology and mechanism of action of riluzole, Neurology

47 (1996) S233eS241.

The authors wish to thank Rottapharm Biotech S.r.l. Monza, Italy

for ﬁnancial support. Dr Francesco Makovec and Prof Fabio Fusi are

kindly acknowledged for fruitful suggestions and encouragement

during the experiments.

Abbreviations

6-OHDA 6-hydroxydopamine

ACSF

AD

Artiﬁcial cerebrospinal ﬂuid

Alzheimer's disease

ALS

CTRL

DAPI

amyotrophic lateral sclerosis

control

4⁰,6-diamidino-2-phenylindole

DCFH-DADichloro-dihydro-ﬂuorescein diacetate

DMF

EGTA

GLU

N,N-dimetilformammide

Ethylene glycol tetraacetic acid

glutamate

H2DCF-DA 2⁰,7'-dichlorodihydroﬂuorescein diacetate

HD

HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

HVA High-voltage-activated

l-NAME L-NG-Nitroarginine methyl ester

Huntington's disease

[16] D. Martin, M.A. Thompson, J.V. Nadler, The neuroprotective agent riluzole

inhibits release of glutamate and aspartate from slices of hippocampal area

CA1, Eur. J. Pharmacol. 250 (1993) 473e476.

[17] M.E. Frizzo, L.P. Dall'Onder, K.B. Dalcin, D.O. Souza, Riluzole enhances gluta-

mate uptake in rat astrocyte cultures, Cell Mol. Neurobiol. 24 (2004) 123e128.

[18] G. Bensimon, L. Lacomblez, V. Meininger, A controlled trial of riluzole in

amyotrophic lateral sclerosis, ALS/Riluzole Study Group, N. Engl. J. Med. 330

(1994) 585e591.

[19] H.M. Bryson, B. Fulton, P. Ben ﬁ eld, Riluzole. A review of its pharmacodynamic

and pharmacokinetic properties and therapeutic potential in amyotrophic

lateral sclerosis, Drugs 52 (1996) 549e563.

[20] R.G. Miller, J.D. Mitchell, M. Lyon, D.H. Moore, Riluzole for amyotrophic lateral

sclerosis (ALS)/motor neuron disease (MND), Cochrane Database Syst. Rev.

(2002) CD001447.

[21] M. Anzini, A. Cappelli, G. Caselli, A. Giordani, F. Makovec, L.C. Rovati,

S. Vomero, Amidine, Thiourea and Guanidine Derivatives of 2-

aminobenzothiazoles and Aminobenzothiazines for Their Use as Pharmaco-

logical Agents for the Treatment of Neurodegenerative Pathologies, in, Google

Patents, 2009.

[22] M. Anzini, A. Chelini, A. Mancini, A. Cappelli, M. Frosini, L. Ricci, M. Valoti,

J. Magistretti, L. Castelli, A. Giordani, F. Makovec, S. Vomero, Synthesis and

biological evaluation of amidine, guanidine, and thiourea derivatives of 2-

amino(6-tri ﬂ uoromethoxy)benzothiazole as neuroprotective agents poten-

tially useful in brain diseases, J. Med. Chem. 53 (2010) 734e744.

[23] Y. Zhu, D. Nikolic, R.B. Van Breemen, R.B. Silverman, Mechanism of inactiva-

tion of inducible nitric oxide synthase by amidines. Irreversible enzyme

inactivation without inactivator modi ﬁ cation, J. Am. Chem. Soc. 127 (2005)

858e868.

[24] M.M. Iravani, L. Liu, S. Rose, P. Jenner, Role of inducible nitric oxide synthase in

N-methyl-d-aspartic acid-induced strio-nigral degeneration, Brain Res. 1029

(2004) 103e113.

[25] R.C. Babbedge, S.L. Hart, P.K. Moore, Anti-nociceptive activity of nitric oxide

synthase inhibitors in the mouse: dissociation between the effect of L-NAME

and L-NMMA, J. Pharm. Pharmacol. 45 (1993) 77e79.

LDH

MEC

MTT

Lactate dehydrogenase

minimal effective concentration

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium

bromide

MW

Molecular weight

NMDA N-methyl-d-aspartate

nNOS

NO

neuronal nitric oxide synthase

Nitric oxide

OGD/R oxygen/glucose deprivation and reperfusion model

PD Parkinson's disease

RP-HPLC Reversed-phase high-performance liquid

chromatography

ROS

SEM

TEA

THF

TLC

TTx

Reactive oxygen species

standard error of measurement

triethylamine

tetrahydrofuran

thin layer chromatography

Tetrodotoxin

Appendix A. Supplementary data

Supplementary data associated

with

this

article

[26] S.G. Kim, H.J. Kim, C.H. Yang, Thioureas differentially induce rat hepatic

microsomal epoxide hydrolase and rGSTA2 irrespective of their oxygen

radical scavenging effect: effects on toxicant-induced liver injury, Chem. Biol.

Interact. 117 (1999) 117e134.

[27] F. Audiau, C. James, Medicaments Containing 2-benzothiazolamine De-

rivatives, Compounds and Their Preparation, in, EP0374041 A1, 1990.

[28] D.R. Shridhar, K. Srinivasa Rao, Synthesis of some new N-Benzothiazinyl-N'-

aryl/pyridyl-thioureas as potential anthelmintic agents, Indian J. Chem. 20B

(1981) 471e473.

(Figures S1eS23, X-Ray crystallography studies in Figure S24eS25,

and the elemental analysis reported in Table S1) can be found, in

the online version, at http://dx.doi.org/10.1016/j.ejmech.2016.1 1.

053.

References

[29] D.R. Shridhar, C.V. Reddy Sastry, O.P. Bansal, P. Pulla Rao, Synthesis & bio-

logical activity of some new N-(2H-1,4-benzoxazin- & benzothiazin-3-yl)-N'-

arylureas, Indian J. Chem. 25B (1986) 874e876.

[30] A.D. Jordan, C. Luo, A.B. Reitz, Ef ﬁ cient conversion of substituted aryl thioureas

to 2-aminobenzothiazoles using benzyltrimethylammonium tribromide,

J. Org. Chem. 68 (2003) 8693e8696.

[1] http://epthinktank.eu/2013/07/17/neurodegenerative-diseases-in-the-

workplace last access March 2015.

[2] R. Djaldetti, N. Lev, E. Melamed, Neuroprotection in progressive brain disor-

ders, Isr. Med. Assoc. J. 5 (2003) 576e580.

[3] A. Nieoullon, Neurodegenerative diseases and neuroprotection: current views

and prospects, J. Appl. Biomed. 9 (2011) 173e183.

630

A. Mancini et al. / European Journal of Medicinal Chemistry 126 (2017) 614e630

[31] L. Farkas, E. Kasztreiner, F. Andrasi, J. Borsi, J. Kosary, E. Csanyi, S. Elek, Thia-

zolylurea derivatives and process for the preparation thereof, (1976) Patent

GB 1437895, 1431976e1437806-1437803. EC: C1437807D1437417/1437804;

C1437807D1437417/1437814.

[32] A. Krishna, M. Biryukov, C. Trefois, P.M. Antony, R. Hussong, J. Lin,

M. Heinaniemi, G. Glusman, S. Koglsberger, O. Boyd, B.H. van den Berg,

D. Linke, D. Huang, K. Wang, L. Hood, A. Tholey, R. Schneider, D.J. Galas,

R. Balling, P. May, Systems genomics evaluation of the SH-SY5Y neuroblas-

toma cell line as a model for Parkinson's disease, BMC Genomics 15 (2014)

1154.

[38] A. Urbani, O. Belluzzi, Riluzole inhibits the persistent sodium current in

mammalian CNS neurons, Eur. J. Neurosci. 12 (2000) 3567e3574.

[39] J. Magistretti, L. Castelli, V. Taglietti, F. Tanzi, Dual effect of Zn2 þ on multiple

types of voltage-dependent Ca2 þ currents in rat palaeocortical neurons,

Neuroscience 117 (2003) 249e264.

[40] E. Castagnetti, M. Schlosser, The tri ﬂ uoromethoxy group: a long-range elec-

tron-withdrawing substituent, Chemistry 8 (2002) 799e804.

[41] N. Croce, S. Bernardini, S. Di Cecca, C. Caltagirone, F. Angelucci, Hydrochloric

acid alters the effect of L-glutamic acid on cell viability in human neuro-

blastoma cell cultures, J. Neurosci. Methods 217 (2013) 26e30.

[33] S. Brogi, S. Butini, S. Maramai, R. Colombo, L. Verga, C. Lanni, E. De Lorenzi,

S. Lamponi, M. Andreassi, M. Bartolini, V. Andrisano, E. Novellino, G. Campiani,

M. Brindisi, S. Gemma, Disease-modifying anti-Alzheimer's drugs: inhibitors

of human cholinesterases interfering with beta-amyloid aggregation, CNS

Neurosci. Ther. 20 (2014) 624e632.

[34] Y.M. Ding, J.D. Jaumotte, A.P. Signore, M.J. Zigmond, Effects of 6-

hydroxydopamine on primary cultures of substantia nigra: speci ﬁ c damage

to dopamine neurons and the impact of glial cell line-derived neurotrophic

factor, J. Neurochem. 89 (2004) 776e787.

[35] C.S. Huang, J.H. Song, K. Nagata, J.Z. Yeh, T. Narahashi, Effects of the neuro-

protective agent riluzole on the high voltage-activated calcium channels of rat

dorsal root ganglion neurons, J. Pharmacol. Exp. Ther. 282 (1997) 1280e1290.

[36] O.P. Hamill, A. Marty, E. Neher, B. Sakmann, F.J. Sigworth, Improved patch-

clamp techniques for high-resolution current recording from cells and cell-

free membrane patches, P ﬂ ugers Arch. 391 (1981) 85e100.

[37] L. Castelli, M.J. Nigro, J. Magistretti, Analysis of resurgent sodium-current

expression in rat parahippocampal cortices and hippocampal formation,

Brain Res. 1163 (2007) 44e55.

[42] C. Brizi, C. Santulli, M. Micucci, R. Budriesi, A. Chiarini, C. Aldinucci, M. Frosini,

Neuroprotective effects of castanea sativa mill. Bark extract in human neu-

roblastoma cells subjected to oxidative stress, J. Cell Biochem. 117 (2016)

510e520.

[43] A. Contartese, M. Valoti, F. Corelli, S. Pasquini, C. Mugnaini, F. Pessina,

C. Aldinucci, G. Sgaragli, M. Frosini, A novel CB2 agonist, COR167, potently

protects rat brain cortical slices against OGD and reperfusion injury, Phar-

macol. Res. 66 (2012) 555e563.

[44] F. Aiello, M. Badolato, F. Pessina, C. Sticozzi, V. Maestrini, C. Aldinucci,

L. Luongo, F. Guida, A. Ligresti, A. Artese, M. Allara, G. Costa, M. Frosini,

A. Schiano Moriello, L. De Petrocellis, G. Valacchi, S. Alcaro, S. Maione, V. Di

Marzo, F. Corelli, A. Brizzi, Design and synthesis of new transient receptor

potential vanilloid Type-1 (TRPV1) channel modulators: identi ﬁ cation, mo-

lecular modeling analysis, and pharmacological characterization of the N-(4-

Hydroxy-3-methoxybenzyl)-4-(thiophen-2-yl)butanamide, a small molecule

endowed with agonist TRPV1 activity and protective effects against oxidative

stress, ACS Chem. Neurosci. 7 (6) (2016) 737e748.

Article Doi

Synthesis and biological evaluation of a new class of benzothiazines as neuroprotective agents

DOI: 10.1016/j.ejmech.2016.11.053

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.ejmech.2016.11.053

Full text of DOI:10.1016/j.ejmech.2016.11.053

Products guided by the article

R&D Labs maybe for 1134963-85-1

Relevant to this article

Hot Product