Pyrazolo[1,4]diazepines as non-peptidic probes of the oxytocin and vasopressin receptors

Article abstract of DOI:10.1016/j.tetlet.2014.06.022

An improved synthesis of differently substituted pyrazolo[1,4]diazepine compounds is reported. In addition, we have used this methodology to obtain non-peptidic compounds to probe the oxytocin and vasopressin receptors.

Full text of DOI:10.1016/j.tetlet.2014.06.022

Tetrahedron Letters xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Pyrazolo[1,4]diazepines as non-peptidic probes of the oxytocin
and vasopressin receptors
Tristan A. Reekie a, Iain S. McGregor b, Michael Kassiou a,c,d,
⇑
a School of Chemistry, The University of Sydney, Sydney, NSW 2006, Australia
b School of Psychology, The University of Sydney, Sydney, NSW 2006, Australia
c Brain and Mind Research Institute, Sydney, NSW 2050, Australia
d Faculty of Health Sciences, The University of Sydney, Sydney, NSW 2006, Australia
a r t i c l e i n f o
a b s t r a c t
Article history:
An improved synthesis of differently substituted pyrazolo[1,4]diazepine compounds is reported. In addi-
tion, we have used this methodology to obtain non-peptidic compounds to probe the oxytocin and vaso-
pressin receptors.
Received 4 April 2014
Revised 20 May 2014
Accepted 5 June 2014
Available online xxxx
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
Pyrazolo[1,4]diazepines
WAY 267,464
Oxytocin
Vasopressin
The neuropeptide oxytocin has been implicated in multiple
central nervous system functions in mammalian species. Increased
levels have been reported to improve trust, alleviate symptoms
related to autism and social phobias, and reduce social anxiety.1
Drug discovery programs have developed novel non-peptidic oxy-
tocin receptor agonists with greater affinity for the receptor and
increased half-life and blood brain barrier permeability. Two of
the first-generation non-peptide agonists reported were WAY
267,464 (1)2 and compound 23 (Fig. 1). However, further studies
with WAY 267,464 have shown that it is not a selective oxytocin
receptor (OTR) agonist as it exhibits only weak in vitro affinity
for the receptor.4 In contrast, strong vasopressin 1A receptor
(V1AR) affinity was observed.4 Recent in vivo results suggest that
V1AR antagonism may be causing the observed biological effects
rather than OTR agonism, or potentially a combination of both.5
To fully investigate the role of OTR or V1AR in specific diseases
non-peptidic molecules are needed to probe these systems.
Figure 1. Oxytocin mimetics WAY 267,464 (1), compound 2 and incorporated head
group pyrazolo[1,4]diazepine 3.
A key feature of compounds 1 and 2 is the pyrazolo[1,4]diazepine
head group 3, a five-step synthesis of which has been previously
reported.3,6 We recently reported an improved synthesis of this head
group and used the new methodology to synthesize heteroatom
analogues of the benzodiazepine center.7 Herein, we report the
exploration of this methodology for its applicability toward the
synthesis of new analogues, namely appending different functional
groups within and around the aromatic moiety. In addition, we also
reporta modified synthesisof compounds 1 and 2 with methodsthat
are operationally simpler and applicable to gram scale. Greater
availability of compounds 1,
analogues allows for further testing to better understand the role
2 and pyrazolo[1,4]diazepine
⇑
Corresponding author. Tel.: +61 2 9351 2745; fax: +61 2 9351 3329.
E-mail address: michael.kassiou@sydney.edu.au (M. Kassiou).
http://dx.doi.org/10.1016/j.tetlet.2014.06.022
0040-4039/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Reekie, T. A.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.06.022

2
T. A. Reekie et al. / Tetrahedron Letters xxx (2014) xxx–xxx
Scheme 1. Synthesis of compound 3. Reagents and conditions: (i) 5 (3.0 equiv), KOH (2.0 equiv), DMF, 100 °C, 1 h; (ii) Pd/C (5 mol %), H2 (1 atm), MeOH, rt, 18 h.
of oxytocin and vasopressin systems in specific diseases and should
aid the development of targeted therapies.
a Zn/H+ reducing system to thereby afford compound 9e, albeit
in a reduced yield of 62%.
Our revised method7 for the synthesis of compound 3 involves a
base-promoted nucleophilic aromatic substitution on pyrazole 4
with 2-nitroaniline (5) to afford compound 6, which can then
undergo a one-pot tandem reductive cyclization to provide the
head group 3 (Scheme 1). This procedure is applicable to the syn-
thesis of compound 3 on a large scale (0.1 mol) and respectable
yields (57% from pyrazole 4).
With these promising results we next tested the applicability of
these reaction conditions with 2-methyl-6-nitroaniline (10)
(Scheme 3). The steric bulk of the methyl substitution adjacent
to the nucleophilic nitrogen was shown to have some impediment
to the reaction, which required a longer reaction time (18 h vs 1 h)
in addition to a reduced yield (44%) of substituted pyrazole 11
being obtained. The 1H and 13C NMR spectra indicated that com-
pound 11 was synthesized as a mixture of atropisomers in a
1:1.5 ratio. The lack of free rotation did not appear to influence
greatly the subsequent reductive cyclization with benzodiazepine
12 isolated in 82% yield.
The final variation to test this methodology was to attempt the
reactions using 1,2-aminonitropyridines (Scheme 4). Reactions
employing both 3-nitropyridin-2-amine (13a) and 2-nitropyridin-
3-amine (13b) proceeded well in the first instance to provide
substituted pyrazoles 14a (45%) and 14b (51%), respectively. How-
ever, when compound 14a was subjected to the tandem reductive
cyclization, neither benzodiazepine 15a, nor any other discernable
products were obtained despite complete consumption of the
starting material. There was a potential for multiple side reactions
to occur if the pyridine nitrogen reacted with the aldehyde group
before reduction. However, a single product was seen via TLC anal-
ysis of the reaction mixture that was no longer present subsequent
to work-up, suggesting instability of the desired product. Support
for this hypothesis was provided when compound 14b was sub-
jected to the same reaction conditions providing the benzodiaze-
pine 15b in 42% yield. Despite the low yield, enough compound
was obtained for characterization, however upon standing, both
at room temperature and at 4 °C, the compound decomposed into
indiscernible products. This suggests that while pyridine analogues
can be synthesized they are unstable under ambient conditions.
The synthesis of WAY 267,464 (1) from benzodiazepine 3
(Scheme 5) initially followed the method of Hibert et al.,6 whereby
amide formation with acid chloride 16 gave compound 17 (76%).
Due to the large number of commercially available 2-nitroani-
lines with additional substituents we thought it appropriate to
elaborate further upon this methodology to determine the scope
of the reactions reported. As part of this investigation, a variety
of compound 3 analogues would also be produced that could then
be elaborated to provide analogues of compounds 1 and 2.
The most common substitution pattern for additionally func-
tionalized 2-nitroanilines that are commercially available is para
to the amine moiety. Accordingly, the initial attempts to expand
upon our synthetic methodology were to react pyrazole 4 with
the 2-nitroanilines 7a–e, which possess an assortment of func-
tional groups (Scheme 2). The initial nucleophilic aromatic substi-
tution with an extra methyl substitution proceeded well to afford
the substituted pyrazole 8a (R = Me) (69%). Substituting the
aromatic ring with electron-withdrawing (R = COOMe) or elec-
tron-donating (R = OMe) groups did not impede the reaction
though a decrease in yield (61% vs 42%) was noted for the latter
case. Halide substitution (R = F or Cl) was also tolerated in the reac-
tion with similar yields (59% and 58%, respectively).
With the substituted pyrazoles 8a–e in hand, we turned our
attention to the tandem reductive cyclization. Using the
palladium-catalyzed reducing conditions outlined previously the
benzodiazepines 9a–d were all obtained in yields P90%. The con-
version of compound 8e into 9e (R = Cl) under these conditions
proved to be somewhat problematic due to the propensity of the
chloride to cleave under the reaction conditions to instead afford
benzodiazepine 3. This problem could be circumvented by using
Scheme 2. Synthesis of substituted benzodiazepines 9a–e. Reagents and conditions: (i) 7 (3.0 equiv), KOH (2.0 equiv), DMF, 100 °C, 1 h; (ii) Pd/C (5 mol %), H2 (1 atm), MeOH,
rt, 18 h. aReaction required Zn (20 equiv), HCl (10 equiv of a 6.0 M aq solution), MeOH, rt, 1 h.
Please cite this article in press as: Reekie, T. A.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.06.022

T. A. Reekie et al. / Tetrahedron Letters xxx (2014) xxx–xxx
3
Scheme 3. Synthesis of benzodiazepine 12. Reagents and conditions: (i) 10 (3.0 equiv), KOH (2.0 equiv), DMF, 100 °C, 18 h; (ii) Pd/C (5 mol %), H2 (1 atm), MeOH, rt, 18 h.
Scheme 4. Efforts toward pyridine-based benzodiazepines. Reagents and conditions: (i) 13a,b (3.0 equiv), KOH (2.0 equiv), DMF, 100 °C, 1 h; (ii) Pd/C (5 mol %), H2 (1 atm),
MeOH, rt, 18 h.
Scheme 5. Revised synthesis of WAY 267,464 (1). Reagents and conditions: (i) 16 (1.3 equiv), Et3N (2.0 equiv), CH2Cl2, rt, 18 h; (ii) Raney Ni (1.0 equiv), H2 (1 atm), MeOH
(saturated with NH3), 65 °C, 18 h; (iii) 19 (1.2 equiv), CDI (1.2 equiv), i-Pr2EtN (2.4 equiv), DMF, rt, 18 h.
Scheme 6. Revised synthesis of amine 20 and the synthesis of compound 2. Reagents and conditions: (i) (COCl)2 (1.1 equiv), DMF (cat.), then 22 (1.1 equiv), Et3N (2.0 equiv),
CH2Cl2, rt, 16 h; (ii) Lawesson’s reagent (1.5 equiv), chlorobenzene, 150 °C (lw), 2 h; (iii) HBr (33 wt % in AcOH), rt, 3 h; (iv) 18 (1.0 equiv), 20 (1.2 equiv), CDI (1.2 equiv),
i-Pr2EtN (2.4 equiv), DMF, rt, 18 h.
Despite many attempts at different amide forming reactions with
the parent carboxylic acid, the only conditions from which a prod-
uct could be obtained were through the corresponding acid chloride
16. The reduction of the aromatic nitrile moiety as previously
reported using a CoCl2/NaBH4 system, while successful on a large
scale, resulted in reduced yields (60%) and required a laborious
and time-consuming work-up procedure. In an attempt to circum-
vent this issue catalytic hydrogenation methods were explored as
an alternative. An atmosphere of hydrogen with Raney cobalt has
previously been shown to reduce selectively aliphatic nitriles into
the corresponding amine,8 however, for the current aromatic nitrile
no reduction took place. The more active Raney nickel had greater
success, though for complete conversion in a reasonable timeframe
(18 h) the reaction did require elevated temperatures. The signifi-
cant benefit of this approach was that filtering the reaction mixture
and subsequent concentration of the solution afforded the target
amine 18 (95%) without further work-up or purification required.
Subjecting amine 18 to urea formation with piperazine 19 (avail-
able in two steps from commercially available material)9 in the
presence of CDI (carbonyldiimidazole), using conditions slightly
modified from those previously reported,6 gave the target WAY
267,464 (1) in 51% yield. Overall, this highly valuable drug target
could be obtained in five steps and 21% yield from a commercially
available starting material using this newly described procedure.
By substituting the secondary amine 19 with pyrrolidine 20,
compound 2 could also be obtained from primary amine 18 using
a similar urea-forming reaction to that of compound 1 (Scheme 6).
A five-step synthesis of pyrrolidine 20 has already been reported,3
but herein we report a new shorter three-step method. The com-
mercially available Cbz protected L-proline 21 was first converted
Please cite this article in press as: Reekie, T. A.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.06.022

4
T. A. Reekie et al. / Tetrahedron Letters xxx (2014) xxx–xxx
into the corresponding acid chloride and then, without isolation,
treated with amine 22 to afford amide 23 as a mixture of rotamers
and conformers (72%). Microwave irradiation of a solution of
amide 23 and Lawesson’s reagent allowed for the fast conversion
of the amide moiety into the corresponding thio derivative
contained within compound 24 (58%), that also appeared as a mix-
ture of rotamers and conformers in the NMR spectra. Cbz deprotec-
tion proved to be more challenging than expected with all
hydrogenolysis methods failing to cleave the protecting group.
Utilizing the previously reported procedure of HBr in acetic acid
afforded the free secondary amine 20 (86%), however, no charac-
terization data had been reported to allow for comparison. Finally,
with compound 20 in hand, it could then be reacted with amine 18
in the presence of CDI to afford non-peptidic compound 2 in 55%
yield.
Acknowledgement
This work has been supported by funding from the National
Health and Medical Research Council (NHMRC).
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2014.06.
022.
References and notes
1. McGregor, I. S.; Callaghan, P. D.; Hunt, G. E. Br. J. Pharmacol. 2008, 154, 358.
2. Rahman, Z.; Resnick, L.; Rosenzweig-Lipson, S. J.; Ring, R. H. WO 2007 050353;
Chem. Abstr. 2007, 146, 455270.
In conclusion, we have expanded upon our methodology for the
synthesis of pyrazolo[1,4]diazepine moieties to incorporate
various functionalities about the aromatic ring. Attempts to incor-
porate a nitrogen atom into the ring were thwarted by the instabil-
ity of the target compounds. Taking the parent benzodiazepine 3,
we were able to improve upon the synthesis of compounds 1 and
2 in terms of practicality and step economy. With this improved
methodology we aim to utilize the substituted benzodiazepine
derivatives to synthesize analogues of compounds 1 and 2 and test
them for activity toward OTR and V1AR. Efforts in this direction are
ongoing and will be reported in due course.
3. Pitt, G.; Batt, A.; Haigh, R.; Penson, A.; Robson, P.; Rooker, D.; Tartar, A.; Trim, J.;
Yea, C.; Roe, M. Bioorg. Med. Chem. Lett. 2004, 14, 4585.
4. Hicks, C.; Jorgensen, W.; Brown, C.; Fardell, J.; Koehbach, J.; Gruber, C. W.;
Kassiou, M.; Hunt, G. E.; McGregor, I. S. J. Neuroendocrinol. 2012, 24, 1012.
5. Hicks, C.; Ramos, L.; Reekie, T.; Misagh, G. H.; Narlawar, R.; Kassiou, M.;
McGregor, I. S. Br. J. Pharmacol. 2014, 171, 2868.
6. Frantz, M.-C.; Rodrigo, J.; Boudier, L.; Durroux, T.; Mouillac, B.; Hibert, M. J. Med.
Chem. 2010, 53, 1546.
7. Reekie, T. A.; Kavanagh, M. E.; Longworth, M.; Kassiou, M. Synthesis 2013, 45,
3211.
8. Reekie, T. A.; Banwell, M. G.; Willis, A. C. J. Org. Chem. 2012, 77, 10773.
9. Hudson, P.; Pitt, G.; Robert, W.; Batt, A. R.; Roe, M. B. WO 2005 023812; Chem.
Abstr. 2005, 142, 280234.
Please cite this article in press as: Reekie, T. A.; et al. Tetrahedron Lett. (2014), http://dx.doi.org/10.1016/j.tetlet.2014.06.022